Cursor
Anysphere
AI first code editor for rapid engineering
30 of 41 features supported
View DetailsDiscover, Compare, and Deploy4,428 Tracked AI Capabilities
The most comprehensive directory to track which AI tools support voice, memory, image generation, and 40+ specific capabilities.
108 AI Tools41 Features4,428 Data Points
Browse all AI toolsShowing 108 of 108 AI tools.
41 of 41 features match your search.
| Feature | Cursor | Midjourney | GitHub Copilot | ChatGPT | ElevenLabs | Claude Code | Notion AI | Canva Magic Studio | Suno | Hugging Face | Sora | Gamma | v0 | Claude | Runway | Stable Diffusion | Figma | Grammarly | Graphify | DeepSeek | Miro AI | Perplexity AI | HeyGen | NotebookLM | Windsurf | Google Gemini | Descript | FLUX | Zapier | Adobe Firefly | Bolt New | DeepL | Whisper | Replit | Udio | Napkin AI | Kling AI | Microsoft Copilot | OpenClaw | Whimsical | LangChain | Julius AI | Google Antigravity | Pika | ClickUp | Ideogram | Lovable | Otter.ai | Eraser | Veo 3 | Warp | Grok | Photoroom | Recraft | Kiro | Replicate | Fireflies.ai | Aider | Linear | Luma Dream Machine | Vertex AI | Adobe Express | Comet | Leonardo.Ai | n8n | Raycast AI | Amazon Bedrock | Fathom | Llama | OpenRouter | Asana | CodeRabbit | Azure OpenAI Service | Groq | Meta AI | QuillBot | Synthesia | InVideo | Together AI | Pinecone | Make | Tabnine | Deepgram | AssemblyAI | Genspark | Jasper | Mistral AI | Wordtune | D-ID | Superhuman | Krea | Stable Audio | Beautiful.ai | You.com | Adobe Podcast | Murf AI | Poe by Quora | Cohere | Writesonic | Fliki | AdCreative.ai | Pi by Inflection | Rytr | Tome | Character.AI | Copy.ai | HuggingChat | HighLevel |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| ✨ Modalities | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
💬 Text Input | Yes. Agent chat is driven by text prompts in the input. You type instructions, attach @ context, and switch models from that same field. | No. Auto populated | Yes | Yes. Core feature | Yes | Yes, requires pro plan. Plain-language prompts drive the agent across the CLI, IDE, desktop, and web. | Yes | Yes. Text prompting is the primary interface. | Yes. Text prompting is the primary interface. | Yes, requires free plan. Hub search, model cards, and Inference Providers take text. HuggingChat is a separate matrix tool. | Yes. Text prompting is the primary interface. | Yes | Yes. Auto populated | Yes. Core feature | Yes. Text prompting is the primary interface. | Yes, requires free plan. Text prompts drive SD 3.5 and the Stable Image generate APIs. | Yes, requires free plan. Canvas, comments, and the Figma agent take prompts. Starter includes limited AI credits. | Yes, requires free plan. You write in the Grammarly editor or any connected surface; it rewrites and proofreads that text. | Yes. Query the code graph in natural language. | Yes. Core feature | Yes. Miro AI generates diagrams, mind maps and summaries from a prompt. | Yes. Core feature | Yes. Text prompting is the primary interface. | Yes, requires free plan. Chat with a notebook to get answers grounded in your uploaded sources. | Yes. Chat and inline prompting throughout the editor. | Yes. Core feature | Yes | Yes, requires free plan. Text prompts drive FLUX text-to-image on the BFL API and open-weight inference. | Yes, requires free plan. Copilot, Agents, Chatbots, and AI fields take natural language. Free Copilot has a daily limit. | Yes. Text prompting is the primary interface. | Yes. Auto populated | No. Auto populated | Partial. Optional prompt steers spelling/style (224 tokens on hosted Groq/OpenAI-style APIs). The model input is audio. | Yes, requires free plan. Replit Chat / Agent 4 turns a prompt into a published app. First prompt is documented as free. | Yes. Text prompting is the primary interface. | Yes. Paste text and it generates matching diagrams and visuals. | Yes, requires free plan. VIDEO 3.0 is prompt-led: text-to-video with scene, dialogue, and camera direction in the prompt. | Yes. Core feature | Yes, requires free plan. You message the agent in WhatsApp, Telegram, Discord, or the desktop chat. It replies in that same thread. | Yes. Whimsical AI generates flowcharts and mind maps from a prompt. | Yes, requires free plan. LangChain/LangGraph agents are text-in. Fleet lets teams build agents in natural language. | Yes. Chat interface over your data. | Yes, requires free plan. Agents are driven by text prompts in the desktop command center, IDE agent panel, and CLI. | Yes. Text prompting is the primary interface. | Yes, requires free plan. Chat, docs, tasks, and Brain² take natural language. Free forever is advertised. | Yes, requires free plan. Text prompts (including JSON prompting on Ideogram 4.0) drive generation in the app and API. | Yes. Auto populated | Yes, requires free plan. Otter Chat and the transcript editor take typed questions over meeting notes. | Yes. DiagramGPT turns a prompt or code into an architecture diagram. | Yes, requires free plan. Text prompts drive Veo 3.1 in Flow, Gemini, and the API; output can include native dialogue and sound. | Yes, requires free plan. Terminal plus Warp Agent / Agent CLI take natural-language tasks. | Yes. Core feature | Partial. Prompts drive staging/backgrounds. It is an editor, not a general chat LLM. | Yes. Text prompting is the primary interface. | Yes, requires free plan. Prompts become specs (requirements, design, tasks) then implementation. | Yes, requires free plan. Every hosted model takes a prompt (or structured input) via the HTTP API and playground. | Yes, requires free plan. AskFred answers questions over your meeting library in the web app. | Yes, requires free plan. You chat in the terminal (or via editor comments) and Aider edits the repo. | Yes, requires free plan. Issues, comments, Linear Asks, and Linear Agent take text. Free includes the agent platform. | Yes. Text prompting is the primary interface. | Yes, requires plus plan. Gemini and other models via Generate Content API and Agent Studio. | Yes, requires free plan. Templates and generative fill take prompts. Distinct from the Firefly web app. | Yes, requires free plan. The sidecar assistant takes natural-language prompts for search, summaries, and page tasks. | Yes. Text prompting is the primary interface. | Yes, requires free plan. Workflows and AI agents take prompts; natural language can become API calls. | Yes, requires pro plan. AI chat in the Raycast launcher accepts text prompts. | Yes, requires plus plan. InvokeModel and Converse APIs accept text prompts across hosted FMs. | Yes, requires free plan. Ask Fathom is typed Q&A over past calls. | Yes. Core feature | Yes, requires free plan. Chat completions via /api/v1/chat/completions. | Yes, requires free plan. Tasks, comments, and AI Teammates take text. Asana Dash is listed. | Yes, requires free plan. Reviews are driven by PR diffs plus agentic chat. Coding Plans accept free-form text, issues, and PRDs. | Yes, requires plus plan. Chat Completions and Responses APIs on deployed OpenAI models. | Yes, requires free plan. Chat completions API and Playground take text messages. | Yes. Core feature | Yes, requires free plan. Paraphraser and grammar tools take typed text. Directory lists Freemium. Homepage was Cloudflare-blocked this round. | No. Auto populated | Yes, requires free plan. Agent Two and storyboard accept text prompts and scripts. | Yes, requires free plan. Serverless inference and playground take chat/completions text. | Partial. Assistant and MCP take text queries. Pinecone is a vector DB, not a general chat LLM. | Yes, requires free plan. Scenario modules and AI/app steps take text and mapped fields. Directory lists Freemium. | Yes, requires plus plan. IDE chat plus code completions. CLI agent on the Agentic Platform. | Partial. Prompts configure Voice Agent behavior. Primary input is audio. | Partial. Prompts steer Speech Understanding. Primary input is audio. | Yes, requires free plan. AI Workspace 6.0 answers in plain English and can query data into dashboards. | Yes. Core feature | Yes. Core feature | Yes, requires free plan. Rewrite, tone, grammar, continue-writing. No credit card required. | Yes, requires plus plan. Scripts and prompts drive talking avatars and Visual Agents. | Yes, requires pro plan. Email compose, docs, and Grammarly Go accept text. Paid product. | Yes. Text prompting is the primary interface. | Yes, requires plus plan. Text prompts drive music and SFX generation. | Yes, requires plus plan. Create with AI: prompt, paste an outline, or attach a document. 14-day trial; directory lists Paid. | Yes. Core feature | No. Enhance Speech is an audio processor, not a text chat LLM. | Yes | Yes. Core feature | Yes | Yes. Core feature | Yes | Yes | Yes. Core feature | Yes | No. tome.app returned 404 on 2026-09-02. The slides product is not serving. Do not treat this as a live writer. | Yes | Yes. Auto populated | Yes | Yes. SMS/Email/Chat |
🖼️ Image Upload / Vision | Yes. You can drag an image into chat or paste a screenshot. Agent also reads image files (.png, .jpg, .gif, .webp, .svg) for vision-capable models. | No. Auto populated | Yes. Vision support in chat | Yes | No | Yes, requires pro plan. Attached photos in the Claude app are passed into the session as vision context. | No | Yes. Upload images for editing, background removal and Magic Edit. | No. No image input. | Yes, requires free plan. Hub covers image/video/audio/3D models and Spaces demos. | Yes. Generate video from a still image. | Yes | Yes. Auto populated | Yes | Yes. Image-to-video and reference images for style. | Yes, requires plus plan. Edit, Control, inpainting, and image-to-image APIs take uploaded images. | Yes, requires free plan. Place images; AI image editing and Weave tools. Agent can add images/files. | No. A writing assistant, not a vision model. | No. No image input. | Yes | Yes. Images can be added to a board and used as context. | Partial. Limited support | Yes. Upload a photo or footage to create an avatar. | Yes, requires free plan. Help documents Gemini Notebook working across text, graphs, and images, and accepting Google Slides and other visual sources. | Yes. Paste screenshots and mockups into the chat as context. | Yes | Yes | Yes, requires plus plan. FLUX.1 Kontext and Fill/Expand endpoints take images for editing. Hosted API is pay-as-you-go. | Unknown. Zaps can move files. A first-party vision model was not on the pricing page. | Yes. Reference images for structure and style matching. | No. Auto populated | No. Auto populated | No. Speech recognition, not vision. | Unknown. Infinite Canvas is a visual design surface. Screenshot/vision upload was not confirmed. | No. No image input. | No. No image input. | Yes, requires free plan. Image-to-video and start/end-frame-to-video are documented for VIDEO 3.0; Omni adds multi-image and video element reference. | Yes | Yes, requires free plan. Users send photos (e.g. a snake in the yard) and the agent identifies them. Vision depends on the model you attach. | No. No image input. | Unknown. Depends on the model you wire. Not a first-party vision SKU. | Yes. Upload charts and images for interpretation. | Unknown. Vision attach in the agent panel was not confirmed on the product homepage or plans doc. | Yes. Image-to-video from a still frame. | Unknown. Whiteboards/clips exist. Vision on Brain² was not specified. | Yes, requires free plan. Remix, Edit, Reframe, Replace Background, and character-reference endpoints take an image. | Yes. Auto populated | No. Meeting transcription, not a vision model. | No. No image input. | Yes, requires free plan. Image-to-video is a documented Veo 3.1 input; the Gemini API model card lists text and image in. | Unknown. Screenshot/vision attach in the agent was not confirmed on the homepage. | Yes | Yes, requires free plan. You upload product photos. Background removal is the core loop. | Yes. Upload reference images for image-to-image and style matching. | Partial. Homepage mentions multimodal inputs and Figma as context. A general image-upload vision API was not separately documented. | Yes, requires plus plan. Many models take reference images (FLUX, Seedream, etc.). You pass files/URLs in the API. | No. Meeting recorder, not a vision model. | Yes, requires free plan. Homepage: add images to the chat for visual context and screenshots. | Unknown. Issue attachments exist. A vision model on Linear Agent was not on the pricing page. | Yes. Image-to-video from a start or end frame. | Yes, requires plus plan. Gemini multimodal accepts images in the API and console. | Yes, requires free plan. You upload photos to edit. That is the editor loop. | Unknown. Vision on the current page is plausible for an in-browser agent, but Comet help does not spell out image attach. | Yes. Image-to-image, inpainting and reference-guided generation. | Unknown. Nodes can fetch files. A first-party vision input was not confirmed. | Partial. Screen capture and file context. General image upload was not separately documented. | Partial. Multimodal models (e.g. Claude on Bedrock) accept images via the Converse API. | No. Meeting notes, not vision. | Yes. Llama 3.2 | Yes, requires plus plan. Homepage: text, images, videos, and audio through one interface. Vision depends on the routed model. | Unknown. Attachments exist. Vision on AI Teammates was not on the homepage. | Unknown. Designs can feed Coding Plans; a general vision-upload path was not confirmed. | Yes, requires plus plan. GPT-4o and later chat models on Azure accept text and images. | Unknown. Vision on hosted models was not confirmed on the quickstart page. | Partial. Limited support | No. Writing assistant, not vision. | No. Auto populated | Yes, requires plus plan. Upload images for clips and storyboards. | Unknown. Vision on hosted models was not confirmed on the marketing homepage. | No. Vector database, not a vision chat product. | Unknown. Homepage timed out this round. File modules likely exist; vision was not confirmed. | Unknown. Vision attach in IDE chat was not confirmed on the pricing page. | No. Speech platform, not vision. | No. Speech platform, not vision. | Unknown. Not confirmed on the public homepage this round. | Partial. Limited support | Yes | No. Writing assistant, not vision. | Yes, requires plus plan. Upload a face photo for talking-photo/video generation. | Unknown. Whether AI reads images in mail was not checked. | Yes. Real-time image-to-image from a sketch or reference. | No. Audio product, not vision. | Yes, requires plus plan. Add custom images; AI image maker; stock library. | Yes | No. Audio enhancement product, not vision. | Yes | Yes | No | Partial. Limited support | No | Yes | No | No | No. No live product at tome.app. | Yes, requires plus plan | No. Auto populated | Yes. With multimodal models | No |
🎨 Image Generation | Yes. Agent can generate images from text or reference images for UI mockups and diagrams. Files land in the project assets/ folder and show inline in chat. | No. Auto populated | No | Yes, requires plus plan. DALL-E 3 | No | No. Documented as a coding agent that edits files and runs commands, not an image generator. | No | Yes. Magic Media generates images from a text prompt. | No. Does not generate still images. | Yes, requires free plan. Diffusers plus hosted image Spaces. The Hub is not a single first-party image model. | No. Does not generate still images. | Yes | Partial. Auto populated | No | No. Does not generate still images. | Yes, requires free plan. Core product: SD 3.5 Large/Turbo/Medium locally, plus hosted Brand Studio and API. | Yes, requires plus plan. Weave tools generate images and vectors. AI credits: 500/mo on Starter, 3,000/mo on Professional Full ($16). | No. Does not generate images. | No. Does not generate still images. | No | Yes. Miro AI can generate images directly onto the board. | No | No. Does not generate still images. | Yes, requires plus plan. Studio can produce infographics and slide decks from sources. Google AI Plus/Pro raise those generation limits; it is not a general art model. | No. Does not generate still images. | Yes | No | Yes, requires free plan. Core product: FLUX.1 [schnell]/[dev] locally, plus FLUX.2 Pro/Max/Flex/Klein via API. | No. You can call an image API as a step. Zapier itself is not an image model. | Yes. Core product, trained on licensed and public-domain content. | No. Auto populated | No. Auto populated | No. Does not generate images. | Unknown. Apps can call image APIs via integrations. A first-party image model was not confirmed. | No. Does not generate still images. | No. Does not generate still images. | Yes, requires free plan. The product homepage lists IMAGE 3.0, text-to-image, image-to-image, and an AI photo editor alongside video. | Yes. DALL-E 3 | No. OpenClaw is a personal-agent harness. It does not ship an image model. | No. Does not generate still images. | No. Framework, not an image model. | No. Produces charts from data rather than generative artwork. | No. Documented as an agentic coding platform, not an image generator. | No. Does not generate still images. | No. You can attach images. ClickUp is not an image model. | Yes, requires free plan. Core product. Docs highlight typography inside the image (logos, POD, layouts). | Partial. Auto populated | No. Does not generate images. | No. Does not generate still images. | No. Veo outputs video with audio, not still images. Still generation in Flow is a different model. | No. A developer terminal/agent platform, not an image generator. | Yes | Yes, requires plus plan. AI backgrounds, shadows, staging. Video is on paid. Credit-based Pro; dollars vary by region. | Yes. Core product. Raster output plus true vector/SVG. | No. A coding agent, not an image model. | Yes, requires plus plan. Hosts FLUX, Imagen, GPT Image, Recraft, and thousands of community image models. Billing is prepaid credit. | No. Does not generate images. | No. A coding agent, not an image model. | No. Issue tracker, not an image model. | No. Does not generate still images. | Yes, requires plus plan. Imagen models in Model Garden and Vertex AI Studio. | Yes, requires plus plan. Firefly-powered generation inside Express. Plan gates were not on the sign-in wall this round. | Unknown. Image models exist on Perplexity web for some plans; not documented as a Comet browser feature. | Yes. Core product, with a library of community and custom models. | No. You can call image APIs via nodes. n8n itself is not an image model. | Unknown. Whether image gen models are in the 32+ model list was not checked. | Partial. Titan Image and third-party image models are in the model catalog. Not a consumer DALL·E-style chat. | No. Does not generate images. | No | Yes, requires plus plan. Unified interface includes image generation across providers. | No. Work management, not an image model. | No. A code-review product, not an image generator. | Yes, requires plus plan. Azure OpenAI lists image generation models in the Foundry catalog. | No. Inference API for language/audio, not an image generator. | Yes | No. Does not generate images. | Partial. Auto populated | Yes, requires plus plan. Agents generate clips and images inside the editor. | Unknown. Together is inference/GPU cloud. Image model hosting was not on the homepage snippet. | No. Does not generate images. | No. You can call an image API as a module. Make itself is not an image model. | No. A coding assistant, not an image generator. | No. Does not generate images. | No. Does not generate images. | Yes, requires plus plan. Team and Enterprise include an AI Image Agent and unlimited zero-credit use of top-tier image models on Team (promo through 31 Dec 2026). Image generation is a named product surface, not a marketing aside. | Yes, requires pro plan | No | No. Does not generate images. | Yes, requires plus plan. FAQ documents a Stable Diffusion portrait generator. | No. Mail and docs product, not image gen. | Yes. Core product, with real-time generation as you draw. | No. Distinct from Stable Diffusion image models. | Yes, requires plus plan. Optional AI-generated images and a consistent AI image style for the deck. | Yes | No. Not an image product. Do not assume Firefly features. | No | Yes, requires plus plan | No | Yes, requires pro plan | Paid Only | Yes | No | Yes | No. Presentation product is gone. Not a live image generator. | Yes, requires plus plan. Generate character images | No. Auto populated | Yes. Stable Diffusion models available | No |
🎙️ Voice Mode | Partial. A microphone in the chat input dictation-transcribes prompts, and iOS can dictate to agents. That is one-way voice input, not a two-way spoken conversation loop. | No. Auto populated | No | Yes, requires plus plan. Advanced voice mode | Yes | Unknown. Voice exists on the Claude consumer app; Code surfaces do not document a voice loop of their own. | No | No. No voice interface. | No. No voice interface. | No. Not a conversational voice assistant. Audio models exist on the Hub. | No. No voice interface. | No | Partial. Auto populated | No | No. No voice interface. | No. Not a conversational voice product. | No. Audio conversations exist for collab. That is not a voice-assistant loop. | No. No conversational voice loop is documented on the plans page. | No. No voice interface. | No | No. No voice interface. | No | No. No voice interface. | Partial. Audio Overviews have an Interactive Mode (beta, English) where you can join the hosts. That is not a full duplex voice chat with the notebook. | No. No voice interface. | Yes | Yes | No. Not a conversational voice product. Docs mention synchronized audio on some FLUX 3 video paths; that is generated soundtrack, not a voice loop. | No. No conversational voice loop is documented on pricing. | No. No voice interface. | Partial. Auto populated | Partial. Auto populated | No. Transcribes speech. It does not talk back as a voice assistant. | No. No conversational voice loop is documented. | No. No voice interface. | No. No voice interface. | No. Native audio is generated on the clip. There is no live conversational voice loop. | Yes | Partial. Community posts wire OpenAI realtime voice to OpenClaw. That is BYOK, not a first-party always-on voice SKU on the homepage. | No. No voice interface. | No. Not a voice-assistant product. | No. No voice interface. | No. No conversational voice loop is documented. | No. No voice interface. | Unknown. AI Notetaker and voice notes are named. A two-way voice loop was not documented. | No. Not a conversational voice product. | Partial. Auto populated | Partial. Live captions and spoken meetings. It is not a two-way voice assistant like ChatGPT Voice. | No. No voice interface. | No. There is no conversational voice loop; you type or attach a frame, then wait for a clip. | No. No conversational voice loop is documented. | No | No. Not a voice assistant. | No. No voice interface. | No. No conversational voice loop is documented. | No. You can run TTS/STT models. That is not a conversational voice loop product. | Partial. Captures spoken meetings. Not a two-way voice assistant. | Yes, requires free plan. Voice-to-code: speak feature requests, tests, or bug fixes. | No. No conversational voice loop is advertised. | No. No voice interface. | Partial. Speech APIs exist. Not a consumer voice-chat loop like Gemini app. | No. Not a voice assistant. | Unknown. Launch copy mentioned voice on mobile; the current Comet product page does not document a voice loop. | No. No voice interface. | No. No conversational voice loop is documented. | Unknown. Voice input was not confirmed on the AI page. | Unknown. Speech models exist on Bedrock. A turnkey voice-chat loop was not checked this round. | Partial. Records spoken meetings. Not a conversational voice assistant. | No | No. Audio generation/transcription may exist via routed models. Not a conversational voice UI. | No. No conversational voice loop is advertised. | No. No conversational voice loop is documented. | Partial. GPT-4o audio models support speech-in/speech-out; that is an API capability, not a consumer voice app. | No. Speech-to-text exists. That is transcription, not a conversational voice loop. | No | No. Not a voice assistant. | No. Auto populated | Unknown. Voiceover features exist; a voice chat loop was not confirmed. | No. Not a conversational voice product. | No. Not a voice assistant. | No. Not a conversational voice product. | No. No conversational voice loop is documented. | Yes, requires plus plan. Voice Agent API is a conversational speech loop. | Yes, requires plus plan. Voice Agent WebSocket is a conversational speech product. | Partial. Workspace 3.0 documents Speakly as Genspark's dictation companion: it turns speech into edited text in any field. That is one-way voice input, not a two-way conversational voice loop. | No | No | No. Not a voice assistant. | Partial. Interactive Visual Agents with speech. Not a general voice assistant. | Unknown. Voice features were not confirmed. | No. No voice interface. | No. Music/SFX generation, not a voice chat assistant. | No. Voice-overs on slides, not a conversational voice assistant. | No | No. Not a conversational voice assistant. | Yes | No | No | No | No | No | Yes | No | No. No live product at tome.app. | Yes, requires plus plan | No. Auto populated | No | Yes. Voice AI Sales System |
🎥 Video Understanding | Partial. Grok Bot accepts video attachments (up to 200 MB) so that Bot can inspect recorded footage. Core Agent chat is documented for images and files, not a general YouTube/video-watch loop, so this is not product-wide video understanding. | No. Auto populated | No. GitHub Copilot assists with code in IDEs, GitHub.com, and the CLI. Official Copilot docs do not describe natively watching or summarizing uploaded video files. | Yes. Via file upload/YouTube | No | Unknown. Not described in the Claude Code overview or mobile client docs. | No | No. Generates and edits video but does not analyse uploaded video. | No. Does not analyse existing video. | Partial. Video models and Spaces exist. Not a first-party video-chat product. | Partial. Uploaded video can be remixed and extended, not analysed or summarised. | No | No. Auto populated | No | Partial. Video is uploaded for editing tools such as inpainting, not for question answering. | No. This listing is the image family, not a video Q&A product. | No. Video in prototypes / Figma Motion. The agent does not watch YouTube as a vision model. | No. Not a video-analysis product. | No. Does not analyse existing video. | No | No. Does not analyse existing video. | No | Partial. Uploaded video is used for translation and lip-sync, not analysis. | Yes, requires free plan. Public YouTube URLs are a supported source type. Video Overviews are a separate studio output with plan limits. | No. Does not analyse existing video. | Yes. Via file upload/YouTube | Yes | No. FLUX 3 docs talk about generating video, not Q&A over uploaded video. | No. Not a video-analysis product. | No. Does not analyse existing video. | No. Auto populated | No. Auto populated | No. May transcribe a video’s audio track. It does not watch/understand video. | No. Not a video-analysis product. | No. Does not analyse existing video. | No. Does not analyse existing video. | Partial. Omni can take a video element as a reference. Credits policy lists video extension as a premium-only function, not a Q&A-on-upload product. | No | Unknown. No first-party video-watch product was on the homepage. A skill could call a video API. | No. Does not analyse existing video. | No. Not a video product. | No. Does not analyse existing video. | No. Not a video-analysis product. | Partial. Existing video can be uploaded to extend or modify, not to summarise. | Unknown. Clips exist. Video-as-vision was not specified. | No. Image generation/editing, not video Q&A. | No. Auto populated | No. Joins Zoom/Meet/Teams for audio. It does not watch the video track as a vision model. | No. Does not analyse existing video. | Partial. Veo can extend clips and take frame direction in Flow/API. Analysing arbitrary uploaded video is Gemini video understanding, not Veo. | No. Not a video-analysis product. | No | No. Can export video on paid. It does not watch/understand arbitrary video. | No. Does not analyse existing video. | No. Not a video-analysis product. | Partial. Hosts caption and video models. Replicate itself does not “watch YouTube” as a chat product. | No. Records call audio. It does not watch the video as a vision model. | No. Not a video-analysis product. | No. Not a video-analysis product. | Partial. Uploaded clips can be extended, not analysed. | Yes, requires plus plan. Gemini video understanding in Agent Studio and the API. | No. Can edit short video. It does not watch/understand arbitrary video as an LLM. | Unknown. Perplexity can transcribe attached video on the web app; Comet-specific video understanding is not documented. | No. Generates short video clips but does not analyse existing video. | No. Not a video-analysis product. | No. Not a video analysis product. | Partial. Video-capable models may be hosted. Bedrock itself is an inference platform, not a YouTube chat product. | No. Audio of the call, not video-as-vision. | No | Partial. Video generation API is documented. Understanding uploaded video was not separately confirmed. | No. Not a video-analysis product. | No. Not a video-analysis product. | Unknown. Video understanding vs Sora video generation was not separated in the catalog page we read. | No. Not a video-analysis product. STT can take video files but only transcribes the first audio track. | No | No. Not a video product. | Partial. Auto populated | Partial. Agents work with video context in the timeline. Not arbitrary YouTube analysis. | No. Not a video-analysis product. | No. Not a video product. | No. Not a video-analysis product. | No. Not a video-analysis product. | No. Audio/speech focus, not video analysis. | No. Audio/speech focus. | Unknown. Not confirmed on the public homepage this round. | No | No | Partial. Homepage: can summarize YouTube videos (and docs/articles/webpages). That is summary, not full video vision. | No. Generates avatar video; does not analyze arbitrary video input. | No. Not a video product. | No. Does not analyse existing video. | No. Not a video product. | No. Can embed/stock video. Does not watch video as a vision model. | No | No. Not a video analysis product. | No | No | No | No | Partial. Processes uploaded video for repurposing into clips, not question answering. | Partial. Analyses uploaded creative assets for performance scoring, not general video. | No | No | No. No live product at tome.app. | No | No. Auto populated | No | No |
🎬 Video Generation | No. Not a video generation product. | No. Not a video generation product. | No. Not a video generation product. | Yes, requires plus plan. Sora video generation is available inside ChatGPT on paid plans. | No. Not a video generation product. | No. Not a video generation product. | No. Not a video generation product. | Yes, requires plus plan. Magic Studio can generate and edit video inside Canva. | No. Not a video generation product. | No. Not a video generation product. | Yes, requires plus plan. OpenAI Sora generates video from text and images; ChatGPT/API access is plan-gated. | No. Not a video generation product. | No. Not a video generation product. | No. Not a video generation product. | Yes, requires plus plan. Gen-3 / Gen-4 text-to-video and image-to-video in the web studio. | No. Not a video generation product. | No. Not a video generation product. | No. Not a video generation product. | No. Not a video generation product. | No. Not a video generation product. | No. Not a video generation product. | No. Not a video generation product. | Yes, requires plus plan. Avatar and stock-to-video generation from a script. | No. Not a video generation product. | No. Not a video generation product. | Yes, requires plus plan. Veo video generation is available in the Gemini app and Flow. | No. Not a video generation product. | No. Not a video generation product. | No. Not a video generation product. | Yes, requires plus plan. Firefly Generate Video creates clips from text and images. | No. Not a video generation product. | No. Not a video generation product. | No. Not a video generation product. | No. Not a video generation product. | No. Not a video generation product. | No. Not a video generation product. | Yes, requires plus plan. Kling text-to-video and image-to-video in the web studio and apps. | No. Not a video generation product. | No. Not a video generation product. | No. Not a video generation product. | No. Not a video generation product. | No. Not a video generation product. | No. Not a video generation product. | Yes, requires plus plan. Pika 2.x text-to-video and image-to-video in the web app. | No. Not a video generation product. | No. Not a video generation product. | No. Not a video generation product. | No. Not a video generation product. | No. Not a video generation product. | Yes, requires plus plan. Google Veo 3 generates video with native audio in Gemini, Flow, and Vertex. | No. Not a video generation product. | No. Not a video generation product. | No. Not a video generation product. | No. Not a video generation product. | No. Not a video generation product. | Partial. Hosts third-party video models via API; it is not a first-party video studio. | No. Not a video generation product. | No. Not a video generation product. | No. Not a video generation product. | Yes, requires plus plan. Dream Machine generates video from text and images. | Yes, requires plus plan. Vertex AI serves Veo and other video generation models. | Unknown. Sign-in wall; Firefly video inside Express was not confirmed this round. | No. Not a video generation product. | Yes, requires plus plan. Motion/video clips from generated stills in Leonardo. | No. Not a video generation product. | No. Not a video generation product. | No. Not a video generation product. | No. Not a video generation product. | No. Not a video generation product. | No. Not a video generation product. | No. Not a video generation product. | No. Not a video generation product. | Yes, requires plus plan. Azure OpenAI catalog includes Sora and a video generation model. | No. Not a video generation product. | No. Not a video generation product. | Unknown. Cloudflare blocked the public site; video generation was not confirmed. | Yes, requires plus plan. Avatar videos generated from a script in the studio. | Yes, requires plus plan. Text-to-video editor with generated and stock footage. | No. Not a video generation product. | No. Not a video generation product. | No. Automation platform; it does not generate video itself. | No. Not a video generation product. | No. Not a video generation product. | No. Not a video generation product. | Yes, requires plus plan. Team and Enterprise include AI Video and all current video generation models (Gemini Veo, Kling, Runway, Seedance, MiniMax Hailuo, PixVerse, Wan, Hunyuan, and more). | No. Not a video generation product. | No. Not a video generation product. | No. Not a video generation product. | Yes, requires plus plan. Talking-head video from a photo or avatar plus a script. | No. Not a video generation product. | Yes, requires plus plan. Krea Video generates and restyles video in the web studio. | No. Not a video generation product. | No. Not a video generation product. | No. Not a video generation product. | No. Audio enhancement product, not a video generator. | No. TTS / voiceover studio, not a text-to-video model. | No. Not a video generation product. | No. Not a video generation product. | No. Not a video generation product. | Yes, requires plus plan. Text-to-video with stock, AI avatars, and generated clips. | No. Not a video generation product. | No. Not a video generation product. | No. Not a video generation product. | No. Not a video generation product. | No. Not a video generation product. | No. Not a video generation product. | No. Not a video generation product. | No. Not a video generation product. |
📁 File Upload (PDF, DOCX) | Yes. Chat @-mentions attach files and folders. Grok Bot also takes PDFs, Office docs, CSV, JSON, source, and notebooks as message attachments. | No. Auto populated | Yes. Via IDE context | Yes | Yes | Yes, requires pro plan. Reads and edits the workspace; mobile Remote Control downloads attached files onto the machine. | Yes, requires plus plan. Via Notion workspace | Yes. Import documents and assets into a design. | No. No document ingestion. | Yes, requires free plan. Upload models, datasets, and Space apps. Dataset viewer is documented. | No. No document ingestion. | Yes | Yes. Auto populated | Yes | Partial. Media uploads only, not documents. | No. Inputs are prompts and images, not PDF/DOCX. | Yes, requires free plan. Import files; Make can add images and files. Not a PDF-chat product. | Unknown. Docs/PDF upload in the editor was not confirmed on the homepage/plans pages checked. | Partial. Indexes a repository rather than accepting ad-hoc uploads. | Yes | Yes. Import documents and assets onto a board. | Yes | Partial. Scripts and media, not general documents. | Yes, requires free plan. PDFs, Google Docs, Google Slides, ePub, copied text, and Drive sources are documented. Pro raises the per-notebook source cap (up to 300). | Yes. Reads files from the open workspace and accepts attachments. | Yes | Yes | No. Inputs are prompts and images, not PDF/DOCX. | Yes, requires free plan. Tables, Forms, and Zap steps move files and structured records. | No. No document ingestion. | Yes. Auto populated | Yes. Auto populated | No. Audio files, not PDF/DOCX. | Yes, requires free plan. Workspaces hold project files. You can import and edit a full codebase in the browser. | No. No document ingestion. | Partial. Text is pasted or imported; not a general document reader. | No. Inputs are prompts, images, and video/element references. PDF/DOCX ingestion is not documented. | Yes | Unknown. Inbox/calendar/skills imply files, but a PDF-chat uploader was not listed on the homepage. | Partial. Import diagrams and images, not general documents. | Partial. You add loaders in code. LangSmith is not a consumer PDF chat. | Yes. Upload CSV, Excel, and other data files for analysis. | Yes, requires free plan. Agents operate on the local workspace/codebase. That is repo context, not a PDF chat uploader. | No. No document ingestion. | Yes, requires free plan. Docs, attachments, and connected Drive/GitHub. | No. Image upload is documented; PDF/DOCX is not. | Yes. Auto populated | Partial. You can import audio/video of meetings. Not a general PDF knowledge base. | Partial. Code and diagram-as-code files, not general documents. | No. No PDF/DOCX ingestion. Inputs are prompts and images (and clip extension), not documents. | Yes, requires free plan. Agents operate on the local (or cloud-agent) codebase. | No | No. Images (and batch). Not PDF/DOCX chat. | No. No document ingestion. | Yes, requires free plan. Works on local workspaces and, on the web, GitHub/GitLab repos in cloud sandboxes. | Partial. API inputs include images, zip datasets for training. Not a PDF chat uploader. | Partial. Upload recordings. Not a general PDF chat product. | Yes, requires free plan. Works on the files in your git repo. That is the codebase, not a PDF uploader. | Yes, requires plus plan. File uploads are unlimited on Basic ($10/user/mo yearly). Free is limited. | No. No document ingestion. | Yes, requires plus plan. RAG with Cloud Storage and document ingestion in Agent Builder. | Partial. Images/video for the editor. Not PDF chat. | Unknown. File attach is documented for Perplexity search, not as a Comet browser control. | No. No document ingestion. | Yes, requires free plan. Workflows handle files, webhooks, and data tables. | Yes, requires pro plan. PDF, CSV, and file context in AI chat. | Yes, requires plus plan. Knowledge Bases ingest documents (S3, etc.) for RAG. | Unknown. Import of recordings was not confirmed on the homepage fetch. | No | Unknown. PDF/DOCX chat was not confirmed on the quickstart page. | Yes, requires free plan. Project docs and attachments in the Work Graph. | Yes, requires free plan. Operates on repository files and PR diffs. That is repo context, not a PDF chat uploader. | Yes, requires plus plan. Files API and Assistants/vector store accept uploaded customer data. | No. Chat is messages. Audio files go to the STT endpoint, not a PDF chat uploader. | No | Unknown. A document uploader was not confirmed (Cloudflare). | Yes. Auto populated | Yes, requires plus plan. Upload media assets to projects. | No. Chat/completions, not a PDF uploader. | Partial. Ingest documents into indexes for RAG. Not a PDF chat UI. | Unknown. Homepage timed out. File/HTTP modules are typical; not confirmed this round. | Yes, requires plus plan. Context Engine connects to Git hosts. Completions use the current file/repo. | Partial. Batch transcription accepts audio files. Not a PDF chat uploader. | Partial. Upload audio/video files for async transcription. | Unknown. Dashboards from data imply files, but a PDF uploader was not specified. | No | No | Partial. Summarizes documents. A PDF-chat workspace was not specified. | Yes, requires plus plan. Upload docs/images for Creative Reality Studio and agents. | Partial. Email attachments and docs. Not a general PDF chat uploader. | No. No document ingestion. | Partial. Audio-to-audio workflows accept reference audio. | Yes, requires plus plan. Attach supporting files to ground Create with AI. Import/export PPTX. | Yes | Partial. Enhance Speech accepts audio uploads when the tool loads. | Yes | Yes | Yes. RAG document upload | No | Yes | Yes | No | No | No. No live product at tome.app. | No | Yes. Auto populated | Yes | Yes |
💻 Code Execution | Yes. Agent runs shell commands in the terminal, watches output, and can execute tests, linters, and builds as part of a task. | No. Auto populated | Yes. Spark/Codespaces integration | Yes, requires plus plan. Advanced Data Analysis | No | Yes, requires pro plan. Runs shell commands, tests, and git operations as part of the agent loop. | No | No. Not a code tool. | No. Not a code tool. | Yes, requires free plan. Spaces run apps (including ZeroGPU). Inference Endpoints serve models. | No. Not a code tool. | No | Yes. Auto populated | No | No. Not a code tool. | No. You run inference; the model does not execute your Python. | Partial. Figma Make and MCP can produce code/PRs. That is codegen, not a Python sandbox in chat. | No. Does not run code. | No. Parses code with static analysis rather than executing it. | Yes | No. Not a code tool. | No | No. Not a code tool. | Yes, requires pro plan. A secure cloud computer can write and run code in the notebook. Google says Ultra and some Workspace tiers have it now; Pro on the web is rolling out. | Yes. Cascade runs terminal commands and tests as part of an agentic loop. | Yes | No | No. You run inference code; the model does not execute your Python. | Partial. Code by Zapier / webhooks on Professional. Not a general Python notebook. | No. Not a code tool. | Yes. Auto populated | No. Auto populated | No. You run inference. The model does not execute your code. | Yes, requires free plan. The IDE runs code, hosts apps, and can keep a background task running. Agent writes production-ready code. | No. Not a code tool. | No. Not a code tool. | No. Not a code-execution product. | No | Yes, requires free plan. Microsoft Execution Containers / sandbox quotes describe running agent code. Skills can execute on your machine. | No. Not a code tool. | Yes, requires plus plan. Deploy path: run agent-generated code in isolated sandboxes. | Yes. Writes and runs Python to analyse data and plot charts. | Yes, requires free plan. The CLI runs shell commands and coding agents. The IDE agent edits and verifies in the project. | No. Not a code tool. | Partial. Codegen Agent is advertised. That is agent code, not a user Python sandbox in chat. | No. Not a code-execution sandbox. | Yes. Auto populated | No. Not a code sandbox. | No. Reads code to draw diagrams but does not execute it. | No. Not a code-execution product. | Yes, requires free plan. The terminal runs commands; agents can execute in the shell and open PRs. | No | No. Not a code sandbox. | No. Not a code tool. | Yes, requires free plan. Agents implement, test, and can run headless in CI. Web sessions run in isolated sandboxes. | No. You call models. Cog packages your code; that is deploy, not a user Python sandbox in chat. | No. Not a code sandbox. | Yes, requires free plan. Automatically lint and test after changes; can fix what those tools report. | Yes, requires free plan. Coding sessions: Linear Agent can set up, run, and test code before returning work (changelog on homepage). | No. Not a code tool. | Partial. Colab Enterprise and notebooks for code. Not a chat Python sandbox. | No. Not a code sandbox. | Unknown. The marketing site shows the assistant building a webpage in the browser; that is not a documented code sandbox. | No. Not a code tool. | Yes, requires free plan. JavaScript or Python anywhere in a workflow; bash scripts; custom nodes. | Unknown. Whether AI runs code locally was not confirmed. | Partial. Agents can call tools and Lambdas. No first-party Python sandbox in chat. | No. Not a code sandbox. | No | No. You implement tools via the Agent SDK. OpenRouter does not sandbox your Python. | No. Not a code sandbox. StackAI workflows are automation, not a notebook. | Partial. Cloud Coding Agent can open PRs. Reviews themselves comment; they do not run your app. | Unknown. Code interpreter on Azure OpenAI was not verified on the catalog page. | Yes, requires free plan. Built-in tools on Compound / gpt-oss include safe code execution on Groq infrastructure. | No | No. Not a code sandbox. | No. Auto populated | No. Video editor, not a code sandbox. | Yes, requires plus plan. Code sandboxes are listed on the product page. | No. Database and inference APIs, not a code sandbox. | Unknown. A custom-code module was not confirmed this round. | Partial. Agentic workflows and CLI can change code, test, and open PRs. Completions-only is not a sandbox executor. | No. STT/TTS APIs, not a code sandbox. | No. STT APIs, not a code sandbox. | Unknown. GenTerminal is listed by name. Sandbox execution was not documented. | No | No | No. Not a code sandbox. | No. Video/avatar platform, not a code sandbox. | No. Not a code sandbox. | No. Not a code tool. | No. Not a code sandbox. | No. Not a code sandbox. | Yes | No. Not a code product. | No | No | No | No | No | No | No | No | No. Never a code sandbox; URL is 404. | No | Partial. Auto populated | No | No |
🔊 Audio Input | Partial. Grok Bot message attachments include audio files (25 MB each) in the same table as documents and images, so pre-recorded audio can be sent for a task. Core Agent chat is not documented as a general audio-file transcribe surface, so this is Grok Bot only. | No. Auto populated | No | Yes | Yes | Unknown. Audio-file input is not documented for Claude Code. | No | No. No audio input. | Yes. Upload audio to extend, cover or remix into a full track. | Yes, requires free plan. Audio models and Spaces. Not a first-party STT studio. | No. No audio input. | No | No. Auto populated | Yes | No. No audio input. | No. No audio-file input is documented for Stable Image. | No. Not a speech-to-text product. | No. No audio-file input is documented. | No. No audio input. | No | No. No audio input. | No | Yes. Upload a voice recording to clone or drive an avatar. | Yes, requires free plan. Audio files are a documented source type. Audio Overviews are generated output, not just playback of uploads. | No. No audio input. | Yes | Yes | No. No audio-file input is documented for image endpoints. | No. Not a speech product. | No. No audio input. | No. Auto populated | No. Auto populated | Yes, requires free plan. Core product: multilingual speech-to-text (and translation to English). | No. No audio-file input is documented. | Yes. Upload audio to extend or remix. | No. No audio input. | Partial. VIDEO 3.0 Omni adds Element Voice Control to bind a character’s voice. That is a reference, not a general audio-file transcribe feature. | No | Unknown. Twilio phone-number demos exist in testimonials. A first-party audio-file transcribe feature was not on the homepage. | No. No audio input. | No. Not a speech product. | No. No audio input. | No. No audio-file input is documented. | No. No audio input. | Partial. AI Notetaker and PRD-from-voice-notes are advertised. | No. No audio-file input is documented. | No. Auto populated | Yes, requires free plan. Core product: live and recorded meeting transcription. Basic includes 300 minutes/month. | No. No audio input. | No. The Veo 3.1 Gemini API card lists text and image inputs only. Native audio is generated on the way out. | No. No audio-file input is documented. | No | No. Not a speech product. | No. No audio input. | No. No audio-file input is documented. | Yes, requires plus plan. Speech and music models are listed (generate speech, music). | Yes, requires free plan. Records Zoom, Meet, Teams and more. Free plan exists; Pro/Business add minutes and integrations. | Partial. Voice-to-code is live mic, not a general audio-file transcriber. | No. Not a speech product. | No. No audio input. | Partial. Speech-to-text models available. Vertex is a platform, not a voice assistant. | Unknown. Audio on video timelines was not confirmed (page was a sign-in wall). | Unknown. Audio-file input is not described in Comet setup or privacy help. | No. No audio input. | No. Not a speech product. You could wire a STT API as a node. | Unknown. Speech input was not checked. | Partial. Speech models are listed. Bedrock is not a dedicated STT consumer app. | Yes, requires free plan. Bot or bot-free meeting capture, then transcripts and summaries. | No | Partial. Audio is listed as a modality. Treat as routed-model support, not a first-party STT studio. | No. Not a speech product. | No. No audio-file input is documented. | Yes, requires plus plan. Audio models cover speech-to-text and GPT-4o audio input. | Yes, requires plus plan. OpenAI-compatible transcriptions/translations. Whisper Large V3 and Turbo. Free-tier 25 MB files. | No | No. Not a speech product. | No. Auto populated | Partial. Voiceover and audio tracks. Not open-ended speech chat. | Unknown. Speech endpoints were not on the homepage snippet. | No. Not a speech product. | No. Not a speech product. | No. No audio-file input is documented. | Yes, requires plus plan. Flux STT and Voice Agent consume live and recorded audio. | Yes, requires plus plan. Realtime and batch STT are the core product. | Unknown. Speakly implies audio. File transcription was not confirmed. | No | No | No. Not a speech product. | Partial. Voice cloning and TTS drive avatars. Not open-ended audio chat. | Unknown. Dictation was not confirmed. | No. No audio input. | Yes, requires plus plan. Audio-to-audio transformation is a listed capability. | Partial. Record or upload voice-overs. Not speech-to-text of meetings. | No | Yes, requires free plan. Enhance Speech is the core product — upload or record audio. | Yes | No | No | No | Yes | No | Yes | No | No. No live product at tome.app. | No | No. Auto populated | No | Yes. Call transcription/AI booking |
🖌️ Canvas / Artifact Editor | Yes. The Agents Window is an agent-first workspace beside the classic IDE editor, with diffs, file search, and split layouts. Agent also opens in a Cmd+I sidepane. | No. Auto populated | Paid Only. Auto populated | Yes. Canvas feature | No | Yes, requires pro plan. Desktop and VS Code surfaces review diffs visually beside the session. | Yes. Auto populated | Yes. The Canva editor canvas is the product. | No. No canvas or artifact panel. | No. Spaces are apps, not a chat canvas. | No. No canvas or artifact panel. | Yes | Yes. Auto populated | Yes. Artifacts feature | Yes. Timeline and canvas editor for composing generations. | Partial. Brand Studio is an edit workspace (inpaint, sketch-to-image). Not a chat canvas. | Yes, requires free plan. The product is a multiplayer canvas. | Partial. The editor is the product. It is not a Claude Artifacts side panel. | Unknown. Not confirmed either way. | No | Yes. The canvas is the product. | No | No. No canvas or artifact panel. | Yes, requires free plan. Studio transforms sources into study guides, briefings, mind maps, slides, and reports — a dedicated output canvas beside chat. | Yes. The editor pane is the shared surface. | Yes. Canvas in Gemini | Yes | Partial. Kontext/Fill/Expand are image editors via API, not a chat canvas. | Unknown. Overlapping columns disagreed (yes vs partial); left unverified rather than copying a no. | Yes. Generative Fill and Expand operate on a canvas. | Yes. Auto populated | No. Auto populated | No. No editor UI. | Yes, requires free plan. Infinite Canvas is a visual design space that applies back onto the app. | No. No canvas or artifact panel. | Yes. Generated visuals are edited directly on a canvas. | Partial. Element Library 3.0 and the studio timeline are a filmmaking canvas, not a chat artifact panel. | Yes | No. Chat/agent, not an artifacts canvas. | Yes. The canvas is the product. | No. Code + LangSmith traces, not a consumer canvas. | Partial. Notebook-style output panel alongside the chat. | Unknown. Overlapping columns disagreed (partial vs yes); left unverified rather than copying a no. | No. No canvas or artifact panel. | Unknown. Overlapping columns disagreed (yes vs partial); left unverified rather than copying a no. | Yes, requires free plan. Edit, remix, reframe, and background tools are first-class in the app and API. | Yes. Auto populated | No. Transcript editor, not a canvas/artifacts UI. | Yes. The canvas is the product. | Partial. Google Flow’s Scenebuilder is the filmmaking canvas around Veo, not a Veo chat artifact panel. | Unknown. Overlapping columns disagreed (no vs partial); left unverified rather than copying a no. | No | Yes, requires free plan. The product is a photo editor canvas (batch, brand kit). | Yes. Infinite canvas for arranging and editing generations. | Unknown. Overlapping columns disagreed (partial vs yes); left unverified rather than copying a no. | No. Playground is a form, not an artifacts canvas. | No. Transcript UI, not a canvas. | Unknown. Overlapping columns disagreed (no vs partial); left unverified rather than copying a no. | No. Issue/diff UI, not an artifacts canvas. | No. No canvas or artifact panel. | No. Studio is a dev console, not a consumer canvas. | Yes, requires free plan. The product is a design canvas. | Yes, requires free plan. The assistant sits in a sidecar beside the page you are browsing. | Yes. Canvas editor for inpainting and outpainting. | Yes, requires free plan. The product is a visual workflow canvas with code nodes beside it. | Unknown. Overlapping columns disagreed (no vs partial); left unverified rather than copying a no. | No. Console playground, not a consumer artifacts canvas. | No. Notes UI, not a canvas. | No | No. Chat/API console, not an artifacts canvas. | No. Lists/boards/timeline, not an artifacts canvas. | Partial. Architectural diagrams and semantic diffs are review surfaces, not a chat canvas. | No. Not a ChatGPT Canvas product. Playgrounds are for testing API calls. | No. Playground is a chat/API console, not an artifacts canvas. | No | No. Text editor, not a canvas. | No. Auto populated | Unknown. Overlapping columns disagreed (yes vs partial); left unverified rather than copying a no. | No. Playground, not a canvas. | No. Database console, not a canvas. | Yes, requires free plan. Visual scenario canvas. | Unknown. Overlapping columns disagreed (no vs yes); left unverified rather than copying a no. | No. Not a canvas product. | No. Not a canvas product. | Unknown. Dashboards exist. A docs/code canvas was not specified. | Yes. Built-in document editor | No | No. Inline rewrite, not a canvas. | Partial. Timeline/storyboard editor for avatar videos. | Partial. Superhuman Docs editor for collaborative writing. | Yes. Live generative canvas with instant feedback. | No. Audio timeline in DAW, not a text canvas. | Unknown. Overlapping columns disagreed (yes vs partial); left unverified rather than copying a no. | No | No. Not a canvas product. | No | No | No. Auto populated | Yes. Article editor | No | Yes | No | No | No. No live canvas. URL 404. | No. Auto populated | Yes. Auto populated | No. Auto populated | No |
🖥️ Screen Sharing / Computer Use | Yes. Cloud Agents run a full desktop and can drive mouse/keyboard plus the browser (computer use). You can take over the remote desktop, then hand control back to the agent. | No. Auto populated | No. Auto populated | Paid Only. Advanced Voice on iOS and Android can share the screen for eligible subscribers. Live Voice does not support screen sharing. | No | Partial. Claude in Chrome can drive a live web app; that is not full-desktop computer use. | No. Auto populated | No. No computer-use capability. | No. No computer-use capability. | No. No computer-use product. | No. No computer-use capability. | No | No. Auto populated | Partial. Computer Use is available in the API (beta), not in the Claude apps. | No. No computer-use capability. | No. No computer-use product. | Partial. Observation mode / audio conversations. Not computer-use of your OS. | No. No computer-use product. | No. No computer-use capability. | No | No. No computer-use capability. | No | No. No computer-use capability. | No. No computer-use / screen-control product. You add sources, you do not hand over the desktop. | No. Operates on the codebase and terminal, not the screen. | Yes | Yes | No. No computer-use product. | No. No computer-use takeover. | No. No computer-use capability. | No. Auto populated | No. Auto populated | No. No computer-use product. | No. No computer-use desktop takeover. Collaboration is inside Replit. | No. No computer-use capability. | No. No computer-use capability. | No. No computer-use / screen-control product. | Yes | Yes, requires free plan. Node/desktop features: a Mac mini with eyes and hands, keyboard and mouse. Computer-use is the pitch. | No. No computer-use capability. | No. No computer-use desktop product. | No. No computer-use capability. | Partial. Browser-in-the-loop agents automate in-page tasks. That is computer-use-like, not a screen-share meeting product. | No. No computer-use capability. | No. No computer-use desktop takeover. | No. No computer-use product. | No. Auto populated | No. No computer-use takeover. | No. No computer-use capability. | No. No computer-use / screen-control product. | Partial. Session links share agent runs. That is not a desktop computer-use product. | No | No. No computer-use product. | No. No computer-use capability. | No. No computer-use desktop takeover is documented. Agents edit the workspace. | No. No computer-use product. | No. No computer-use takeover. | No. No computer-use product. | No. No computer-use takeover of your desktop. | No. No computer-use capability. | No. No computer-use desktop product. | No. No computer-use product. | Yes, requires free plan. The assistant is documented acting on the live page (forms, shopping, posting) rather than only summarizing. | No. No computer-use capability. | No. No computer-use desktop takeover. | Partial. Screen context for AI queries. Not live screen sharing to others. | No. No computer-use desktop product. | No. No computer-use takeover. | No | No. No computer-use product. | No. No computer-use takeover. | No. No computer-use desktop takeover. | Partial. computer-use-preview is listed for the Responses API computer-use tool. That is not a consumer screen-share chat. | No. No computer-use product. | No | No. No computer-use. | No. Auto populated | No. No computer-use product. | No. No computer-use product. | No. No computer-use product. | No. No computer-use takeover. | No. No computer-use product. | No. No computer-use product. | No. No computer-use product. | Unknown. Not confirmed on the public homepage this round. | No | No | No. No computer-use. | No. No computer-use product. | No. No computer-use product. | No. No computer-use capability. | No. No computer-use product. | No. Present live; not computer-use. | No | No. No computer-use product. | No | No | No. Auto populated | No | No | No | No | No | No. No live product at tome.app. | No. Auto populated | No. Auto populated | No. Auto populated | No |
| 🧠 Memory & Context | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
🧠 Persistent Memory | Partial. Grok Bots retain facts and summaries across work. Core Agent chat relies on Rules instead of auto-memory between completions, so this is not ChatGPT-style global memory. | No. Auto populated | No | Yes, requires plus plan | No | Yes, requires pro plan. CLAUDE.md plus auto memory that saves learnings across sessions. | Yes. Embedded in your Notion workspace | No. No cross-session memory of the user. | No. No cross-session memory of the user. | No. The Hub stores repos, not a personal chat memory. | No. No cross-session memory of the user. | Partial. Remembers brand and theme settings across decks, not conversational facts. | Yes. Auto populated | No | No. No cross-session memory of the user. | No. Each generation is a job. Brand ID models are weights, not chat memory. | No. Files persist. The agent is not a personal-facts memory profile. | Yes, requires plus plan. Style guide, brand tones, snippets, and Knowledge Share persist org/voice rules. | No. No cross-session memory of the user. | No | No. No cross-session memory of the user. | No | No. No cross-session memory of the user. | Partial. Sources and chats persist inside a notebook. There is no global cross-notebook user-memory profile like ChatGPT Memory. | Yes. Memories persist project context across sessions. | No | No | No. Each generation is a job. No user memory layer. | Partial. Zaps, Tables, and storage persist. That is automation state, not personal chat memory. | No. No cross-session memory of the user. | Yes. Auto populated | Partial. Auto populated | No. Each run is a job. | Partial. Workspaces persist. That is project state, not ChatGPT-style memory of you. | No. No cross-session memory of the user. | No. No cross-session memory of the user. | No. No cross-session user memory layer is documented. Credits and membership are the persistence story. | No | Yes, requires free plan. Homepage: memories and skills live on your computer and persist; the longer you use it, the more it becomes yours. | No. No cross-session memory of the user. | Yes, requires free plan. Own your harness, context, and memory. LangGraph is built for persistent state. | Unknown. Not confirmed either way. | Partial. Custom agents and ~/.gemini/ plus .agents/ configs persist. That is project/user config, not ChatGPT-style memory of you. | No. No cross-session memory of the user. | Yes, requires free plan. Brain²: infinite memory of tasks, docs, tone, and preferences. Multiplayer AI shares team context. | No. Custom models remember a brand look. That is not chat memory of you. | Yes. Auto populated | Partial. Otter Chat over your meeting library. That is search over transcripts, not a personal-facts memory profile. | No. No cross-session memory of the user. | No. No cross-session user memory. Flow may keep projects; that is the studio, not a Veo memory layer. | Partial. Enterprise lists cross-harness agent memory (research preview). Warp Drive stores snippets. Not ChatGPT memory. | No | No. Brand Kit stores brand assets. That is not chat memory of you. | No. No cross-session memory of the user. | Yes, requires free plan. Steering files (.kiro/steering/, ~/.kiro/steering/, AGENTS.md) persist conventions across sessions and surfaces. | No. Predictions are jobs. You keep state in your app. | Partial. AskFred searches past meetings. Not a personal-facts memory profile. | Partial. Repo map plus git history persist. Optional conventions files. Not a hosted memory of you. | Partial. Issues, projects, and agent context persist. That is work-item state, not a personal chat-memory profile. | No. No cross-session memory of the user. | Partial. RAG corpora and agent session config. Distinct from consumer Gemini memory. | No. Brand kits persist. Not chat memory of you. | Unknown. Max “Brain” memory is documented for Perplexity Computer, not as a Comet default. | No. No cross-session memory of the user. | Partial. Workflows, credentials, and data tables persist. That is automation state, not chat memory of you. | Partial. Chat history in Raycast. No cross-device cloud memory advertised. | Partial. Knowledge Bases and AgentCore session memory. You configure retention. | Partial. Ask Fathom searches past calls. Not a personal-facts memory profile. | No | No. API is stateless aside from your credits/keys. No user memory layer. | Yes, requires plus plan. Shared memory: AI Teammates learn from completed tasks, feedback, and preferences. | Yes, requires plus plan. Learnings persist team preferences. You can also document standards for reviews. | Partial. Responses API and Assistants threads persist message history when you enable those features; models are otherwise stateless. | No. Stateless API. You keep history in your app. | No | No. Not a chat-memory product. | Partial. Auto populated | Yes, requires plus plan. Long-term project memory across Agent Two sessions. | No. Stateless inference. You keep history. | Yes, requires plus plan. Vector indexes persist embeddings for retrieval — core product. | Partial. Scenarios and data stores persist. That is automation state, not chat memory. | Yes, requires pro plan. Context Engine plus Coaching Guidelines persist org standards. Zero code retention otherwise. | Partial. Voice Agent sessions retain context within a call. | Partial. Voice Agent retains session context during a call. | Unknown. Workspace persistence was not specified beyond “workspace” branding. | No | No | No. Not a personal-facts memory profile. | Partial. Visual Agents retain context within a session. Long-term memory not advertised. | Partial. Sounds Like You and writing style learn over time. | No. No cross-session memory of the user. | No. Generations are jobs, not persistent chat memory. | No. Themes and libraries persist. Not chat memory of you. | No | Unknown. Session persistence was not observable. | No | No | No | No | Partial. Retains brand voice and project settings, not conversational memory. | Yes | Yes | No | No. No live product at tome.app. | Yes. Character memory per persona | Yes. Auto populated | No | Yes. Contact CRM memory |
📏 Large Context Window | Yes. Each chat has a fixed model context window with a usage ring. Cloud Agents let you pick context size; several models document up to 1M-token extended context. | No. Auto populated | Yes. Large codebase context | Partial. Standard context | No | Yes, requires pro plan. Indexes the repository; Pro lists a 200k context window for Claude. | Partial. Bound by the underlying model; Notion does not publish a figure. | No. Not a language model, so this does not apply. | No. Not a language model, so this does not apply. | Unknown. Depends on the model you load. No Hub-wide window. | No. Not a language model, so this does not apply. | No | Yes. Auto populated | Yes. 1M+ tokens | No. Not a language model, so this does not apply. | No. Not an LLM context-window product. | No. AI credits, not a published token window. | No. Not an LLM context-window product. Generative prompts are metered (100 free / 2,000 Pro). | No. Not a language model, so this does not apply. | Yes. 64k tokens | No. Not a language model, so this does not apply. | Partial. Standard context | No. Not a language model, so this does not apply. | Unknown. Google publishes source-count limits, not a token window for Gemini Notebook chat. | Partial. Indexes the whole repository; the usable window depends on the selected model. | Yes. 1M+ tokens | No | No. Not an LLM context-window product. | No. Task allowances, not a published LLM context window. | No. Not a language model, so this does not apply. | Yes. Auto populated | No. Auto populated | No. Not an LLM context-window product. | Unknown. Intelligent model routing is documented; no token window is published on pricing. | No. Not a language model, so this does not apply. | No. Not a language model, so this does not apply. | No. Not an LLM context-window product. Clip length is 3–15 seconds on VIDEO 3.0. | Partial. Standard context | Unknown. Window size is the model you bring (ChatGPT sign-in, Claude Max, etc.). OpenClaw does not publish its own token limit. | No. Not a language model, so this does not apply. | Unknown. Depends on the model you choose. No LangChain token window. | Partial. Bound by the underlying model; not published as a headline figure. | Unknown. Uses Gemini agent models; no token window is published on the Antigravity plans page. | No. Not a language model, so this does not apply. | Unknown. No token window is published. Brain² “picks the best model”. | No. Not an LLM context-window product. | Yes. Auto populated | No. Not an LLM context-window product. Limits are minutes, not tokens. | No. Not a language model, so this does not apply. | No. Prompt cap is 1,024 tokens on the Gemini API Veo 3.1 card — not a large LLM context window. | Unknown. You pick models (Claude, GPT, Gemini, etc.). No single window is published. | Partial. Standard context | No. Not an LLM context-window product. | No. Not a language model, so this does not apply. | Unknown. Uses multiple FMs via Bedrock/Auto. No token window is published on the pricing page. | Unknown. Depends on the model you run. No single Replicate window. | No. Limits are storage/minutes, not an LLM token window. | Unknown. Depends on the LLM you connect. No window is published as an Aider product spec. | No. Issue limits (250 on Free), not an LLM token window. | No. Not a language model, so this does not apply. | Unknown. Depends on the Gemini/model version deployed. | No. Not an LLM context-window product. | Unknown. No token window is published for the Comet assistant. | No. Not a language model, so this does not apply. | Unknown. Depends on the model you connect. No n8n token window is published. | Unknown. Depends on the selected model (32+ providers). | Unknown. Depends on the foundation model you invoke. No single Bedrock window. | No. Not an LLM token-window product. | Yes. 128k tokens | Yes, requires plus plan. Window is whatever the selected model publishes. Catalog is at openrouter.ai/models. | No. Not an LLM token-window product. | Unknown. No token window is published on the pricing page. | Yes, requires plus plan. Depends on the deployed model; gpt-chat-latest lists a 400k context window. | Unknown. Depends on the hosted model. No single Groq window is on the quickstart page. | Partial. Standard context | Unknown. Word limits were not fetched this round. | No. Auto populated | Unknown. Agent context limits were not published. | Unknown. Depends on the open model you host. No single Together window on the homepage. | No. Retrieval layer, not an LLM context window. | No. Operations quota, not an LLM token window. | Unknown. Depends on the connected LLM. No window is published as a Tabnine spec. | Unknown. Agent context limits were not on the homepage snippet. | Unknown. Agent context limits were not checked. | Unknown. No token window is published on the homepage. | Partial. Standard context | Partial. Standard context | Unknown. No token window on the homepage. | Unknown. Agent context limits were not published on the homepage. | Unknown. Depends on underlying models. Not published on homepage. | No. Not a language model, so this does not apply. | No. Not an LLM chat product. | Unknown. No token window on the homepage. | Partial. Depends on which underlying model the query is routed to. | No. Audio processor, not an LLM. | No | Partial. Depends on which model you pick; Poe routes to many providers. | Yes. 128k tokens | Partial. Standard context | No | No | Partial. Holds long conversations, but no context size is published. | No | No. No live product at tome.app. | Partial. Long chats are summarised as they grow rather than kept in full. | Yes. Auto populated | Yes. Depends on selected model | Partial. Depends on the model configured for the workflow. |
⚙️ Custom Instructions | Yes. Project and user Rules (.cursor/rules plus User Rules) inject persistent instructions into Agent. Always Apply, glob, and @-mention rules are documented. | No. Auto populated | Yes. Via .github/copilot-instructions.md | Yes | No | Yes, requires pro plan. Project CLAUDE.md, path-scoped rules, and --append-system-prompt steer behavior. | Yes | No. No global instruction setting. | No. No global instruction setting. | No. No global chat instruction profile for the Hub. | No. No global instruction setting. | No | No. Auto populated | Yes | No. No global instruction setting. | Partial. Brand Studio Campaigns and Brand ID models steer style. That is not a global chat system prompt. | Partial. Skills, Make kits, and design-system libraries. Not a global chat instruction card. | Yes, requires plus plan. Style guide and brand tones steer suggestions. Enterprise can manage features centrally. | No. No global instruction setting. | No | No. No global instruction setting. | No | No. No global instruction setting. | Unknown. A global custom-instructions setting was not confirmed in the help articles checked. | Yes. Workspace rules files steer the assistant per project. | Yes. Via Gems (custom AI personas) | No | No. No global instruction profile. Fine-tunes/LoRAs are weights, not a system prompt. | Yes, requires free plan. Zap logic, filters, paths, and agent/chatbot instructions. | No. No global instruction setting. | No. Auto populated | No. Auto populated | Partial. A short prompt can guide spelling. Not a global instruction profile. | Unknown. Plan mode is on Core. A global instruction profile was not confirmed on the homepage. | No. No global instruction setting. | No. No global instruction setting. | No. No global instruction profile. Each job is a prompt plus optional references. | No | Yes, requires free plan. Persona onboarding, skills, and cron/heartbeats are the instruction layer. | No. No global instruction setting. | Yes, requires free plan. Prompts, graphs, and Fleet natural-language agent specs. | Unknown. Not confirmed either way. | Yes, requires free plan. Custom Agents are Markdown+YAML under .agents/agents/ or ~/.gemini/config/agents/. | No. No global instruction setting. | Yes, requires free plan. Personality, sprint methods, and preference memory are the instruction layer. | Partial. Magic Prompt and custom/on-brand models steer style. Not a global chat instruction profile. | Yes. Auto populated | Unknown. Custom vocabulary/vocabulary was not re-checked on the 2026 pricing page. | No. No global instruction setting. | No. No global instruction profile for Veo. Each job is a prompt plus optional frames. | Yes, requires free plan. Factory config as code, autonomy settings, and data controls steer agents. | No | Partial. Brand Kit / style presets. Not a global chat instruction profile. | No. No global instruction setting. | Yes, requires free plan. Steering markdown (always / fileMatch / manual) plus AGENTS.md. | Partial. System prompts / input schemas per model. Not a global user profile. | Unknown. Custom vocabulary was not re-checked on the pricing table. | Yes, requires free plan. Config and conventions (comments in-editor, aider config) steer how it edits. | Yes, requires free plan. Triage rules, templates, and agent workflows. | No. No global instruction setting. | Yes, requires plus plan. System instructions in Agent Studio and API prompts. | Unknown. Brand kits likely. Not confirmed behind the sign-in wall. | Unknown. A global assistant-instruction setting was not found in Comet help. | No. No global instruction setting. | Yes, requires free plan. Workflow logic, guardrails, and human-in-the-loop steps are the instructions. | Partial. Extensions and prompts customize behavior. No global system profile documented. | Yes, requires plus plan. Guardrails, system prompts, and agent instructions in the console/API. | Unknown. Not specified on the homepage. | Yes | Yes, requires free plan. System messages plus org data policies that restrict providers. | Yes, requires plus plan. Create your own AI Teammates; scoped permissions in the same console as humans. | Yes, requires free plan. Quiet/Chill/Assertive profiles, path instructions, and team standards configure review style. | Partial. System prompts are API parameters, not a ChatGPT-style global custom-instructions UI. | Yes, requires free plan. System messages in the chat API. | No | Unknown. Modes (standard/fluency) exist in market; not re-verified this round. | No. Auto populated | Yes, requires plus plan. Brand kits and agent instructions per project. | Yes, requires free plan. System prompts in the API. | Partial. Assistant system prompts. Not a global user profile. | Yes, requires free plan. Scenario logic, filters, and routers are the instructions. | Yes, requires plus plan. Coaching Guidelines and governance controls define boundaries and behavior. | Yes, requires plus plan. Configure Voice Agent prompts and behavior. | Yes, requires plus plan. Configure Voice Agent prompts and guardrails. | Unknown. Not confirmed on the public homepage this round. | No | No | Partial. Formal/casual tone and rewrite modes. Not a global instruction profile. | Yes, requires plus plan. Configure Visual Agent personality and knowledge. | Partial. Sounds Like You adapts tone. Not a full system-prompt editor. | No. No global instruction setting. | Partial. Prompt and style controls per generation. | Yes, requires plus plan. Themes, brand control, image-style prompts, language (100+). | Yes | No. Not an LLM with custom instructions. | No | Yes | Yes | No | No | No | No | Yes | No. No live product at tome.app. | Yes. Character persona definition | Yes. Auto populated | Yes. System prompt support | Yes. AI Employee configuration |
📜 Conversation History | Yes. Past chats persist so you can @Chats into a new prompt. CLI resume/continue reloads prior threads; iOS keeps an agent inbox of running conversations. | No. Auto populated | Yes | Yes | Yes | Yes, requires pro plan. Sessions persist across terminal, IDE, web, and mobile, including VS Code history. | Yes | No. Not a chat product, so there is no conversation to configure. | No. Not a chat product, so there is no conversation to configure. | No. Not a chat-history product. (HuggingChat is separate.) | No. Not a chat product, so there is no conversation to configure. | Yes | Yes. Auto populated | Yes | No. Not a chat product, so there is no conversation to configure. | No. Not a chat-history product. | Yes, requires free plan. Comments, version history (30 days on lower plans), agent on-canvas chat. | Partial. Documents and suggestion history live in the product. Not a multi-thread chat log like ChatGPT. | No. Not a chat product, so there is no conversation to configure. | Yes | No. Not a chat product, so there is no conversation to configure. | Yes | No. Not a chat product, so there is no conversation to configure. | Yes, requires free plan. Notebooks persist. Cross-app sync keeps the same notebook in the Gemini app and the standalone product. | Yes. Cascade conversations are saved per workspace. | Yes | Yes | No. Dashboard jobs are not a chat history product. | Yes, requires free plan. Zap run history. Agents/Chatbots keep their own threads. | No. Not a chat product, so there is no conversation to configure. | Yes. Auto populated | No. Auto populated | No. Not a chat product. | Yes, requires free plan. Chat with the Agent persists in the workspace. | No. Not a chat product, so there is no conversation to configure. | No. Not a chat product, so there is no conversation to configure. | Unknown. The studio keeps generated work, but a searchable chat-history product was not confirmed. | Yes | Yes, requires free plan. Chat history lives in the apps you already use plus local memory. | No. Not a chat product, so there is no conversation to configure. | Yes, requires plus plan. LangSmith traces and datasets persist agent runs. | Yes. Chat threads are saved and can be resumed. | Yes, requires free plan. Antigravity 2.0 groups conversations into Projects and can schedule messages. | No. Not a chat product, so there is no conversation to configure. | Yes, requires free plan. Chat, comments, and docs persist. | Partial. The web app keeps generated images you can remix. That is a gallery, not a chat log. | Yes. Auto populated | Yes, requires free plan. Meetings and chats are saved in your Otter account. | No. Not a chat product, so there is no conversation to configure. | No. Veo is a generator, not a chat log. Conversational video editing is documented for Gemini Omni, not Veo. | Yes, requires free plan. Cloud conversation storage (limited on Free, unlimited on Build). Session links for sharing. | Yes | No. Project/export history, not chat threads. | Partial. Generation history is kept per project, but there is no chat thread to resume. | Yes, requires free plan. IDE, CLI, web, and Crew keep sessions. Web sandboxes keep running after you close the laptop. | Partial. Playground and prediction logs. The API is stateless per call. | Yes, requires free plan. Meetings and AskFred threads persist in the account. | Yes, requires free plan. Terminal chat plus git commits are the history. You can undo with git. | Yes, requires free plan. Issue comment history and activity. | No. Not a chat product, so there is no conversation to configure. | Partial. Session state in agents. API calls are otherwise stateless. | No. Project files, not chat threads. | Partial. Comet is a full browser with history (cleared in Incognito). Assistant thread history across devices is not documented. | No. Not a chat product, so there is no conversation to configure. | Yes, requires free plan. Execution logs and workflow history (days vary by plan). | Yes, requires pro plan. AI chat history persists in Raycast. | Partial. AgentCore sessions persist within configured limits. Stateless per InvokeModel call. | Yes, requires free plan. Call library persists. | No | Yes, requires free plan. The site is a chat UI over models. API history is your client’s job. | Yes, requires free plan. Task comments and project activity persist. | Yes, requires free plan. PR review threads and in-app agentic chat persist on the Git host and CodeRabbit UI. | Partial. Optional Responses API / Assistants threads store history in your resource; chat completions are otherwise stateless. | Partial. Playground sessions exist. The API is stateless unless you resend messages. | No | Unknown. Not confirmed this round. | No. Auto populated | Yes, requires plus plan. Agent chat history persists within projects. | Partial. Playground/console history. The API is stateless. | Partial. Assistant sessions may retain history. API is index-oriented. | Yes, requires free plan. Execution history of scenarios. | Yes, requires plus plan. IDE chat is a conversation surface. Auditability of usage is listed. | Partial. Call logs and transcripts in console. API is per-session. | Partial. Transcripts stored per job. Realtime is per-session. | Unknown. The Chrome extension keeps conversation history for the current session so you can ask follow-ups. A persistent saved-chat product was not documented. | Yes | Yes | Unknown. Not specified on the homepage. | Partial. Agent chat logs in studio. API is per-generation. | Yes, requires pro plan. Email threads and docs history persist in the product. | No. Not a chat product, so there is no conversation to configure. | Partial. Web app may show past generations. API is per-call. | No. Decks, not chat threads. | Yes | Unknown. History feature was not observable. | Yes | Yes | Yes | Yes | No | Yes | Yes | No | No. No live product at tome.app. | Yes | Yes. Auto populated | Yes | Yes. Full lead history |
| 🔌 Integrations | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
🌐 Web Search / Browsing | Yes. Agent has a Web tool that generates search queries and performs live web searches during a task. | No. Auto populated | Yes. Bing web search in chat | Yes | No | Yes, requires pro plan. Paid Claude plans include web search; Code also pulls live context through MCP servers. | Yes, requires plus plan | No. No browsing capability. | No. No browsing capability. | No. You search the Hub, not the live web as an LLM tool. | No. No browsing capability. | Yes | Yes. Auto populated | Yes. Built-in web search since 2025 | No. No browsing capability. | No. Does not browse the live web. | Unknown. Visual search of your files is listed. Live web search was not. | No. Does not browse the live web to answer questions. | No. No browsing capability. | Yes | No. No browsing capability. | Yes. Real-time focus | No. No browsing capability. | Yes, requires free plan. Discover sources via Fast Research (web or Drive) and Gemini Deep Research inside Notebook, which can browse many sites and import the report. | Yes. Can fetch documentation and web results as context. | Yes | No | No. Does not browse the live web. | Unknown. Agents can use apps that search. A built-in cited search was not listed. | No. No browsing capability. | Yes. Auto populated | No. Auto populated | No. Does not browse the web. | Unknown. A first-party live-web search toggle was not confirmed. | No. No browsing capability. | No. No browsing capability. | No. Does not browse the live web. | Yes. Real-time focus | Unknown. Skills can browse. A built-in cited search box was not on the homepage. | No. No browsing capability. | Partial. You add a search tool. Not a built-in cited search box. | Unknown. Not confirmed either way. | Partial. Frontend pitch is browser-in-the-loop agents. That is agent browsing, not a cited search box. | No. No browsing capability. | Yes, requires plus plan. Brain² “checks your workspace, apps, and the web.” | No. Does not browse the live web. | Yes. Auto populated | No. Searches your meetings, not the live web. | No. No browsing capability. | No. Does not browse the live web. | Unknown. A first-party live-web search toggle was not confirmed. | Yes. Real-time focus | No. Does not browse the web. | No. No browsing capability. | Unknown. MCP can pull tools; a first-party live-web search toggle was not confirmed. | No. Not a search engine. A model might browse if the author built that. | No. Searches your conversations, not the live web. | Partial. You can add web pages to the chat as context. That is not a live search box. | No. Not a web-search product. | No. No browsing capability. | Partial. Grounding with Google Search in Gemini API. Enterprise Agent Platform feature set. | No. Not a search engine. | Yes, requires free plan. Comet is built around Perplexity’s answer engine in the address bar and assistant. | No. No browsing capability. | Partial. You can add search/HTTP nodes. Not a built-in cited search box. | Yes, requires pro plan. Web search with inline references in AI chat. | Partial. Agents can wire a search tool (e.g. via Lambda). Not a built-in cited search box. | No. Searches your calls, not the live web. | No | Unknown. Some routed models search. A first-party search toggle was not on quickstart. | No. Not a web-search product. | Unknown. A first-party live-web search toggle was not confirmed. | Unknown. Grounding/search on Azure OpenAI was not checked. Distinct from Copilot. | Yes, requires free plan. Built-in tools include web search and website visiting on Compound / gpt-oss. | No | No. Not a search engine. | No. Auto populated | Unknown. Whether agents browse the web was not checked. | No. Not a search product. | No. Retrieves from your index, not the open web. | Unknown. HTTP modules can call search APIs. Not a built-in cited search. | Unknown. MCP can reach external services. A first-party web search toggle was not confirmed. | Unknown. Whether Voice Agent can search the web was not checked. | Unknown. Web search in Voice Agent was not confirmed. | Unknown. Genspark started as search-like. Current Workspace 6.0 search behavior was not detailed. | No | No | Partial. “We check at least 5 sources before deeming a fact credible.” That is grounded rewrite, not a search box. | Unknown. Whether agents can browse the web was not checked. | Unknown. Web search in AI features was not confirmed. | No. No browsing capability. | No. Not a search product. | Partial. Optional web search when generating, per FAQ. | Yes. Core feature, real-time | No. Not a search product. | No | Yes | Yes. Real-time web connector | No | Yes | Yes | No | No | No. No live product at tome.app. | No | Yes. Auto populated | Yes. DuckDuckGo/Brave search | No |
🔌 Plugins / Extensions | Yes. Cursor Marketplace installs official plugins that bundle rules, skills, MCP servers, and hooks. Teams can ship a private marketplace; VS Code extensions still work in the editor. | No. Auto populated | Yes. VS Code extensions | No. Replaced by GPTs | No | Yes, requires pro plan. MCP servers and the plugin system connect Drive, Jira, Slack, and custom tools. | Yes. Via Notion integrations | No. No plugin marketplace. | No. No plugin marketplace. | Yes, requires free plan. Libraries (Transformers, Diffusers, smolagents) plus Spaces MCP listings. | No. No plugin marketplace. | No | No. Auto populated | No | No. No plugin marketplace. | Partial. ComfyUI, Replicate, Fireworks, and DeepInfra are documented hosts. That is not a Stability plugin store. | Yes, requires free plan. Community plugins on Starter; private plugins on higher plans. Generative plugins (beta). | Yes, requires free plan. Works inside other apps (homepage shows Zendesk). That is an editor overlay, plus a browser extension. | No. No plugin marketplace. | No | Yes. Marketplace of apps and integrations. | No | No. No plugin marketplace. | No. No third-party plugin marketplace. Sources and Google apps are the integration surface. | Partial. MCP servers extend it, rather than a first-party plugin marketplace. | No | No | Partial. Community hosts FLUX on Replicate/Hugging Face. That is not a BFL plugin marketplace. | Yes, requires free plan. 9,000+ app connections. Premium apps need Professional. | No. No plugin marketplace. | No. Auto populated | No. Auto populated | Partial. Community wraps it in many apps. OpenAI’s repo is the model, not a plugin store. | Yes, requires free plan. 100+ integrations (OpenAI, Stripe, Google Workspace, and more). | No. No plugin marketplace. | No. No plugin marketplace. | No. No third-party plugin marketplace is documented for Kling itself. | No | Yes, requires free plan. Community skills (/autoreview, /handoff, etc.). Growing skill ecosystem is the product pitch. | No. No plugin marketplace. | Yes, requires free plan. Model providers, tools, and 30+ Agent Server endpoints. | No. No plugin marketplace. | Yes, requires free plan. Codelab: Antigravity IDE supports local and remote MCP servers; a blog covers IDE Extensions. | No. No plugin marketplace. | Yes, requires free plan. 100+ products, GitHub/Figma/Salesforce, MCP, automations. | Yes, requires plus plan. Official docs list MCP alongside the app and API. Developer site also exposes an MCP server for the API docs. | No. Auto populated | Partial. Zoom/Meet/Teams plus Zapier and an MCP server. Not a GPT-style plugin store. | No. No plugin marketplace. | No. No third-party plugin marketplace for Veo itself. | Yes, requires free plan. Any model/harness, Slack/Linear/GitHub mentions, BYOK, and factory automations from GitHub/webhooks. | No | Partial. REST API and MCP. Not a GPT plugin store. | No. No plugin marketplace. | Yes, requires free plan. MCP, Agent Client Protocol, Open VSX extensions, Skills.md, and 500+ CLI tools. | Yes, requires free plan. Thousands of community models, client libraries, webhooks, MCP server. | Yes, requires pro plan. Integrations (CRM, etc.) on paid; Zapier listed. Not a GPT plugin store. | No. No plugin marketplace. It is a CLI you script. | Yes, requires free plan. Git, Slack intake, MCP, API/webhooks. Salesforce is an add-on. | No. No plugin marketplace. | Yes, requires plus plan. Extensions, tools, and Model Garden integrations. | Unknown. Not confirmed this round. | Yes, requires free plan. Chromium extensions are supported; Incognito disables them unless you turn them back on. | No. No plugin marketplace. | Yes, requires free plan. 500+ integrations, custom nodes, MCP, HTTP/GraphQL. | Yes, requires pro plan. Raycast extensions and AI Extensions ecosystem. | Yes, requires plus plan. Action groups, Lambda tools, and third-party integrations for agents. | Yes, requires free plan. Syncs to Slack, Salesforce, HubSpot, Notion, Asana. | No | Yes, requires free plan. Hosted MCP at mcp.openrouter.ai plus Agent SDK tools. | Yes, requires plus plan. MCP and AI connectors; StackAI connects to any app. | Yes, requires plus plan. MCP connections (5 on Essentials, more on higher plans), plus Jira and Linear. | Partial. Functions and tools are documented; there is no ChatGPT plugin store. | Yes, requires free plan. Remote MCP tool calling, plus Python/JS SDKs and AI SDK provider. | No | Unknown. Integrations were not confirmed this round. | No. Auto populated | Partial. Stock media and integrations in the editor. | Partial. Endpoints, batch, dedicated clusters. Not a GPT plugin store. | Yes, requires plus plan. MCP server, integrations, and rerank models. | Yes, requires free plan. Thousands of app modules (directory heritage). Count was not re-fetched. | Yes, requires plus plan. Works in major IDEs. Agentic plan adds MCP (Jira, Confluence, Docker, CI). | Partial. Integrations and webhooks for voice workflows. | Partial. LLM Gateway and integrations connect upstream models. | Unknown. Claw, Speakly, GenClipboard, GenTerminal are listed. A plugin store was not. | No | No | No. No plugin store. | Partial. Canva, PowerPoint, and Google Slides integrations. | Yes, requires pro plan. 800+ tool integrations in the Superhuman suite. | No. No plugin marketplace. | Partial. DAW plugin integrates into production workflows. | No. No plugin store. | Yes. Apps | No. Not a plugin marketplace product. | No | Yes. Third-party bots | Yes. Data connectors | No | No | Yes | No | No | No. No live product at tome.app. | No | No. Auto populated | No | Yes. App marketplace |
🔧 API Access | Yes. Cloud Agents API, TypeScript/Python SDKs, Admin API, Analytics API, and related endpoints are documented. They run agent workflows, not a raw chat-completions API. | No. Auto populated | Yes, requires pro plan | Yes, requires pro plan | Yes | Partial. The Agent SDK and Console API keys automate Claude Code; there is no separate public Code REST product besides the Anthropic API. | Yes, requires pro plan | Yes. Connect APIs for apps and asset workflows. | Unknown. Not confirmed either way. | Yes, requires plus plan. Inference Providers (45k+ models, no service fees listed) and Inference Endpoints from $0.033/hour. | Unknown. Not confirmed as generally available through the API. | Paid Only | Yes. Auto populated | Yes, requires pro plan | Yes. Video generation API for developers. | Yes, requires plus plan. Stable Image REST APIs (generate, upscale, edit, control) on platform.stability.ai. Trial credits, then paid. | Yes, requires pro plan. REST APIs, webhooks, MCP (20 tool calls/month on Starter; more on paid). | Partial. Enterprise lists an Audit Logs API. That is not a public generate-text REST API. | Unknown. Not confirmed either way. | Yes, requires pro plan | Yes. REST API and developer platform. | Yes, requires pro plan | Yes. Video generation API for developers. | No. Gemini Notebook is a consumer/Workspace product. There is no public NotebookLM generate API documented in the help center. | No. An IDE product, not an API. | Yes, requires pro plan | Yes | Yes, requires plus plan. Official async API at api.bfl.ai (plus eu/us regional). Auth is an x-key. Pay-as-you-go, no seat fee. | Yes, requires pro plan. Webhooks, REST, and Zapier MCP. MCP calls consume tasks. Free is limited. | Yes, requires enterprise plan. Firefly Services API for enterprise customers. | Partial. Auto populated | Paid Only. Auto populated | Yes, requires plus plan. Hosted by OpenAI, Groq, and others via transcriptions APIs. The GitHub repo is local inference. | Unknown. You build apps with APIs. A public Replit generate API was not confirmed. | Unknown. Not confirmed either way. | Unknown. Not confirmed either way. | Yes, requires pro plan. Kling publishes an Open Platform (document-api / global/dev). API billing is prepaid resource packages, not the consumer membership. | Yes, requires pro plan | Unknown. Gateway/node is local. A public REST product SKU was not on the marketing homepage. | Unknown. Not confirmed either way. | Yes, requires free plan. OSS libraries plus LangSmith / Agent Server APIs. | Unknown. Not confirmed either way. | Yes, requires free plan. Antigravity SDK is a documented Python surface for custom agents on the harness. | Unknown. Not confirmed either way. | Yes, requires plus plan. API calls are listed among product capabilities. | Yes, requires plus plan. Public REST API at api.ideogram.ai, billed separately from app subscriptions. Playground is in the API Dashboard. | Partial. Auto populated | Unknown. An Otter developer API was not confirmed on the pricing page this round. | Unknown. Not confirmed either way. | Yes, requires pro plan. Gemini API Veo 3.1 is paid-tier only. Legacy veo-3.0 IDs shut down 30 June 2026. | Partial. Enterprise Analytics API is listed. Agent CLI is the main programmatic surface, not a public generate REST API. | Yes, requires pro plan | Yes, requires pro plan. REST API for listings plus an MCP server. | Yes. REST image generation API, billed by credits. | Partial. Headless CLI in CI/CD and ACP into other IDEs. That is not a public REST generate API. | Yes, requires free plan. HTTP API, Node/Python clients, deployments. Get started for free; pay for compute. | Yes, requires pro plan. API is listed on Business in the public pricing comparison. | No. Aider consumes LLM APIs. It is not a hosted Aider REST generate API. | Yes, requires free plan. API, webhooks, and MCP access are listed on the pricing comparison. | Yes. Dream Machine API for developers. | Yes, requires plus plan. Vertex AI REST/gRPC SDK. $300 credits for new accounts. | Unknown. Not confirmed on the public URL this round. | No. Comet is a browser. Programmatic access is the separate Perplexity API, not a Comet REST product. | Yes. Production image generation API on paid tiers. | Yes, requires free plan. Control n8n via API and CLI. Trigger from webhooks. | No. Consumer launcher, not a public inference API. | Yes, requires plus plan. AWS SDK, InvokeModel, Agents, and Knowledge Bases APIs. Usage billing. | Unknown. A public developer API was not on the homepage. | Yes. Free via API providers | Yes, requires plus plan. OpenAI-compatible API. Credits, no seat subscription. Client SDKs and OpenAI SDK drop-in. | Unknown. A public developer API is long-standing; it was not re-confirmed on the 2026 homepage. | Yes, requires enterprise plan. API access is listed on Enterprise. Lower plans are Git-host/IDE/CLI. | Yes, requires plus plan. REST API for deployed models in your Azure resource. Usage billing. | Yes, requires free plan. api.groq.com OpenAI-compatible chat and audio. Create a key on console.groq.com. | No | Unknown. Not confirmed this round. | Paid Only. Auto populated | Unknown. Public API was not confirmed on homepage. | Yes, requires free plan. Inference API, fine-tuning, GPU cloud. Usage billing, not a seat. | Yes, requires plus plan. REST/gRPC index and inference APIs. | Yes, requires plus plan. Webhooks and an API to manage scenarios are the usual Make surface. Details not re-fetched. | No. You may bring your own LLM endpoint. Tabnine is not a public generate REST API. | Yes, requires plus plan. REST and WebSocket APIs for STT, TTS, and Voice Agent. | Yes, requires plus plan. REST and WebSocket APIs for STT and Voice Agent. | Yes, requires plus plan. Team and Enterprise admin dashboards include an API Keys page to view and revoke keys created by team members, so members can call Genspark from code on those plans. | Yes, requires pro plan | Yes, requires pro plan | Unknown. Not on the homepage. | Yes, requires plus plan. Creative Reality and Agents HTTP APIs. | No. Consumer productivity app, not a public API. | Unknown. Not confirmed either way. | Yes, requires plus plan. Stable Audio API for music/SFX generation. | Unknown. Not on the homepage. | Yes, requires pro plan | Unknown. Public API was not confirmed this round. | Yes | Yes, requires pro plan | Yes | Yes, requires pro plan | Paid Only | Paid Only | No | No | No. No live API. tome.app is 404. | No | Paid Only. Auto populated | Yes. Via Hugging Face Inference API | Yes. Open API |
⚡ Zapier Integration | Unknown. No first-party Cursor Zapier app was listed on cursor.com/docs. Automations document GitHub, GitLab, Slack, Linear, and webhooks instead. | No. Auto populated | No | Yes | No | Unknown. No first-party Zapier listing was checked. | Yes | Yes. Official Zapier integration. | Unknown. Not confirmed either way. | Unknown. No first-party Zapier app was checked. | Unknown. Not confirmed either way. | No | No. Auto populated | No | Unknown. Not confirmed either way. | Unknown. No first-party Zapier app was checked. | Unknown. Not checked this round. | Unknown. No first-party Zapier app was checked. | Unknown. Not confirmed either way. | No | Yes. Official Zapier integration. | No | Yes. Official Zapier integration. | Unknown. No first-party Gemini Notebook Zapier app was checked. | Unknown. Not confirmed either way. | No | Yes | Unknown. No first-party Zapier app was checked. | Yes, requires free plan. This is Zapier. Native Zaps are the product. | Unknown. Not confirmed either way. | No. Auto populated | No. Auto populated | Unknown. No first-party Zapier app was checked. | Unknown. No first-party Zapier app was checked. | Unknown. Not confirmed either way. | Unknown. Not confirmed either way. | Unknown. No first-party Kling Zapier app was checked. | No | Unknown. No first-party Zapier listing was checked. | Unknown. Not confirmed either way. | Unknown. No first-party Zapier listing was checked. | Unknown. Not confirmed either way. | Unknown. No first-party Zapier app was checked. | Unknown. Not confirmed either way. | Unknown. Integrations exist. A first-party Zapier listing was not checked. | Unknown. No first-party Zapier app was checked. | No. Auto populated | Yes, requires plus plan. Zapier is listed on paid/Basic materials as a meeting-notes automation path. | Unknown. Not confirmed either way. | Unknown. No first-party Veo Zapier app was checked. | Unknown. No first-party Zapier app was checked. | Yes. Zapier lists Grok by xAI as an official app. Zaps can Ask Grok, categorize text, extract fields from text or images, and generate images from a prompt. | Unknown. No first-party Zapier listing was checked. | Unknown. Not confirmed either way. | Unknown. No first-party Zapier app was checked. | Unknown. No first-party Zapier listing was checked. | Yes, requires pro plan. Zapier is listed among Fireflies integrations. | No. No Zapier integration is documented. | Unknown. A Zapier listing was not checked this round. | Unknown. Not confirmed either way. | Unknown. No first-party Zapier listing was checked. | Unknown. Not checked. | Unknown. No first-party Comet Zapier app was checked. | Unknown. Not confirmed either way. | No. n8n is a Zapier alternative, not a native Zapier app. Use n8n nodes instead. | Unknown. Zapier was not checked. | Unknown. No first-party Zapier listing was checked this round. | Unknown. No Zapier listing was checked this round. | No | Unknown. No first-party Zapier app was checked. | Unknown. Not checked this round. | Unknown. No first-party Zapier app was checked. | Unknown. No first-party Zapier listing was checked. | Unknown. No first-party Zapier app was checked. | No | Unknown. Not checked. | No. Auto populated | Unknown. Zapier was not checked. | Unknown. No first-party Zapier listing was checked. | Unknown. Zapier integration was not checked this round. | No. Make is a Zapier alternative, not a native Zapier app. | Unknown. No first-party Zapier app was checked. | Unknown. Zapier listing was not checked. | Unknown. Zapier listing was not checked. | Unknown. Zapier has no genspark app listing (zapier.com/apps/genspark returns 404), so an official Genspark Zap was not confirmed. | Yes | No | Unknown. Not checked. | Unknown. Zapier listing was not checked. | Unknown. Zapier integration was not checked. | Unknown. Not confirmed either way. | Unknown. Zapier was not checked. | Unknown. Not checked. | No | Unknown. Zapier was not checked. | No | No | No | Yes | No | No | No | No | No. No live product to integrate. | No | Yes. Auto populated | No | Yes. Premium Zapier app |
🤖 Custom GPTs / Bots | Yes. Grok Bots are named durable agents with a profile, skills, routines, and shareable templates. You create, duplicate, and share a Bot the way Custom GPTs are specialized agents. | No. Auto populated | Paid Only. Auto populated | Yes. Custom GPTs / GPT Store | No | Partial. Skills and sub-agents package workflows; they are not a Custom GPT store. | Paid Only. Auto populated | No. No agent builder; Magic Studio features are fixed tools. | No. No agent or bot builder. | Yes, requires free plan. Spaces and smolagents let you publish agents/apps. | No. No agent or bot builder. | No | No. Auto populated | Partial. Projects carry custom instructions and files, but there is no public bot directory. | No. No agent or bot builder. | No. No Custom GPT store. Fine-tunes are weights. | Yes, requires plus plan. Generative plugins: describe a plugin and the agent builds it. | No. No Custom GPT store. Superhuman Go agents are a separate Superhuman surface. | No. No agent or bot builder. | No | No. No agent builder. | No | No. No agent or bot builder. | No. You do not publish Custom GPT-style bots. A notebook is a private (or shared) source collection. | Partial. Workflows and rules rather than shareable custom agents. | Yes. Gems (custom AI personas) | No | No. No Custom GPT store. Fine-tunes are weights. | Yes, requires free plan. Chatbots and Agents are standalone add-ons with basic access on Free. | No. No agent or bot builder. | No. Auto populated | No. Auto populated | No. Not a bot store. | No. You publish apps, not a Custom GPT store. | No. No agent or bot builder. | No. No agent or bot builder. | No. No Custom GPT-style Kling agents. | Paid Only | Yes, requires free plan. You run your own claw with skills and persona; share with a team. | No. No agent or bot builder. | Yes, requires free plan. You build agents (Fleet in NL, or code) and deploy them. | No. No agent or bot builder. | Yes, requires free plan. Custom Agents (Markdown) can be selected in the GUI or run via agy --agent. | No. No agent or bot builder. | Yes, requires free plan. Build your own agent; Super Agent 24/7. | No. Custom models are image weights, not shareable chat bots. | No. Auto populated | No. No custom-GPT store. | No. No agent or bot builder. | No. No Custom GPT-style Veo agents. | Partial. Factories and harness choice specialize agents. Not a GPT Store. | No | No. No custom-GPT store. | No. No agent or bot builder. | Yes, requires free plan. Custom agents plus steering; ACP into other tools. Kiro Crew is the open-source coordinator. | No. You deploy models, not a GPT store. | No. No custom-GPT store. | No. No bot store. | Yes, requires free plan. Build/deploy agents; Linear Agent is included on Free. | No. No agent or bot builder. | Yes, requires plus plan. Agent Builder and Agent Studio deploy custom agents. | No. No custom-GPT store. | Partial. Email Assistant and Background Assistants are Max-tier agents, not a public custom-GPT store. | No. No agent or bot builder. | Yes, requires free plan. You build AI agents as workflows and expose them (e.g. chat/webhook). | Partial. Custom AI Extensions act like specialized bots. | Yes, requires plus plan. Bedrock Agents and AgentCore are deployable custom agents. | No. No custom-GPT store. | No | No. No GPT Store. Agent SDK is for your own agents. | Yes, requires plus plan. Pre-built AI Teammates plus create-your-own. | Partial. Learnings and path instructions specialize the reviewer. Not a GPT Store. | Partial. Assistants / Agents APIs let you build assistants; there is no public Custom GPT store. | No. No GPT Store. You build agents on the API. | No | No. No custom-GPT store. | No. Auto populated | Partial. Agent Two is a project-scoped video agent. | No. No GPT store. | Partial. Pinecone Assistant is a RAG bot over your data. | No. You build scenarios, not a GPT store. | Partial. Coaching Guidelines specialize agents. Not a GPT Store. | Yes, requires plus plan. Voice Agent is a deployable speech bot. | Yes, requires plus plan. Voice Agent is a deployable speech bot. | Yes, requires free plan. Workspace 3.0 introduces Genspark Claw as your first AI employee: a persistent agent with custom skills and scheduled workflows that acts across email, calendar, and messaging — not just a one-off chat. | No | No | No. No custom-GPT store. | Yes, requires plus plan. Visual Agents are deployable custom avatars. | No. Not a GPT store. Distinct from Grammarly matrix tool. | No. No agent or bot builder. | No. Not a bot platform. | No. No custom-GPT store. | No | No. Not a bot platform. | No | Yes. Create and publish custom bots | Yes. Auto populated | No | No | No | No | No | No. No live product at tome.app. | Yes. Auto populated | Yes. Auto populated | Yes. Auto populated | Yes. AI Employee |
👥 Real-time Collaboration | Partial. You can share a Cloud Agent URL with teammates (read-only unless team follow-ups are on). Grok Bot group chats exist. This is not a simultaneous multiplayer editor session. | No. Auto populated | Partial. Auto populated | Partial. Chats can be shared and Projects shared with a team, but there is no live multiplayer session. | Yes | Unknown. Multiplayer editing of one Code session was not confirmed. | Yes. Auto populated | Yes. Multiplayer editing and commenting on a design. | No. Tracks are shared as links rather than co-edited. | Yes, requires free plan. Git-based Hub collaboration is the core pitch. | Unknown. Not confirmed either way. | Yes | No. Auto populated | No | Yes. Shared team workspaces. | No. No multiplayer editor is documented. | Yes, requires free plan. Multiplayer, comments, spotlight, unlimited viewers. | Partial. Team style guides and analytics. Not a multiplayer cursor in one doc. | Unknown. Not confirmed either way. | No | Yes. Multiplayer editing with teammates. | No | Yes. Shared team workspaces on paid plans. | Partial. Notebooks can be shared. Upgrade help says sharing does not change another collaborator’s source limit. Live multiplayer cursor editing is not documented. | Unknown. Not confirmed either way. | Yes | Yes | No. No multiplayer editor. | Yes, requires plus plan. Team: 25 users, shared Zaps and connections. Enterprise unlimited users. | Unknown. Not confirmed either way. | No. Auto populated | No. Auto populated | No. No multiplayer product. | Yes, requires plus plan. Agent 4 multi-user collaboration and Pro collaborator/viewer seats are documented. | No. Tracks are shared as links rather than co-edited. | Unknown. Not confirmed either way. | Unknown. Team spaces share credits. Live multiplayer editing was not documented. | Yes | No. Run it for yourself or share with a team — not a multiplayer canvas. | Yes. Multiplayer editing with teammates. | Partial. Teams share LangSmith. Not a multiplayer chat canvas. | Unknown. Not confirmed either way. | Unknown. Repo-shared .agents/ files help teams. Live multiplayer editing was not documented. | Unknown. Not confirmed either way. | Yes, requires free plan. Multiplayer AI and shared workspace. | No. No multiplayer editor is documented. | No. Auto populated | Partial. Shared meetings and live notes. Not a multiplayer AI document. | Yes. Multiplayer editing with teammates. | Unknown. Flow link sharing exists in changelogs; live multiplayer editing was not confirmed. | Yes, requires plus plan. Warp Drive collaboration (unlimited on Build) and real-time session links so others can watch/steer an agent. | No | Unknown. Team/Brand Kit exists; live multiplayer was not confirmed. | Unknown. Team plans exist; live multiplayer editing not confirmed. | Partial. Team billing and shared steering. Live multiplayer cursors were not documented. | No. No multiplayer canvas. | Partial. Shared meeting workspace. Not a multiplayer AI doc. | No. Single-user CLI. Share via git. | Yes, requires free plan. Shared issues; unlimited members on Free. | Unknown. Not confirmed either way. | Partial. IAM-shared console and notebooks. Not multiplayer chat. | Unknown. Not confirmed this round. | Unknown. Shared multiplayer browsing sessions are not documented. | Unknown. Not confirmed either way. | Yes, requires plus plan. Unlimited users; collaboration, RBAC, and workflow diffs on paid/business plans. | No. Single-user launcher. | No. IAM-shared console, not a multiplayer chat canvas. | Partial. Shared call notes into Slack/CRM. Not a multiplayer AI doc. | No | No. No multiplayer chat. | Yes, requires free plan. Multiplayer: humans and agents on the same plan. | Partial. PR threads are multi-human. Not a live pair-programming canvas. | Unknown. Team sharing in Azure AI was not checked. | No. No multiplayer chat. | No | No. Not a multiplayer doc. | No. Auto populated | Yes, requires plus plan. Multiplayer editing on the same project. | No. No multiplayer chat. | No. Shared org access, not multiplayer editing. | Unknown. Team editing was not confirmed this round. | No. Team admin/analytics, not a multiplayer chat session. | No. No multiplayer canvas. | No. No multiplayer product. | Unknown. Not confirmed on the public homepage this round. | Yes. Team collaboration | No | No. Not multiplayer. | Unknown. Team editing was not confirmed. | Yes, requires pro plan. Real-time collaborative docs. | Unknown. Not confirmed either way. | No. No multiplayer editing advertised. | Yes, requires plus plan. Real-time collaboration, permissions, locked themes. | No | Unknown. Collab was not checked. | Yes | No | No. Auto populated | Yes. Team workspaces | No | Yes | No | No | No. No live product at tome.app. | Partial. Auto populated | Partial. Auto populated | No. Auto populated | Yes. Agency-client collaboration |
🐦 Live Social Data Integration | No. No live social firehose; web search is general web, not X/Twitter streams. | No. Not a live social-firehose product. | No. No live social firehose. | No. No live social firehose. | No. No live social firehose. | No. No live social firehose. | No. No live social firehose. | No. Not a live social-firehose product. You can add public web/YouTube URLs as sources. | No. No live social firehose in the Gemini app. | No. No live social firehose. | No. You can connect social apps. That is not a live sentiment firehose. | No. No live social firehose. | No. Not a live social-firehose product. | No. No live social firehose. Community clone-and-try is not a social-data API. | No. No live social firehose. | No. X API access is a partner quote, not a live social-firehose product. | No. No live social firehose product. | No. Not a live social-firehose product. | No. No live social firehose. | No. No live social firehose. | No. No live social firehose. | No. No live social firehose. | No. Not a live social-firehose product. | Yes. Real-time X data | No. No live social firehose. | No. Not a live social-firehose product. | No. No live social firehose. | No. No live social firehose. | No. No live social firehose. | No. No live social firehose. | No. No live social firehose product. | No. Exports to social; that is not a live firehose. | No. Not a live social-firehose product; it can browse social sites like any browser. | No. Not a live social-firehose product. You can add social nodes. | No. No social firehose. | No. No live social firehose product. | No. No live social firehose. | No. No live social firehose. | No. No live social firehose. | No. Not a live social-firehose product. | No. No live social firehose. | No. No live social firehose. | No. No live social firehose. | No. No social firehose. | No. No live social firehose. | No. No social firehose. | No. You can connect social apps. That is not a live sentiment firehose. | No. Not a live social-firehose product. | No. No social firehose. | No. No social firehose. | No. No live social firehose is advertised. | No. No live social firehose. | No. No social firehose. | No. No social firehose. | No. No social firehose. | No. No live social firehose. | No. No social firehose. | No. No live product at tome.app. | ||||||||||||||||||||||||||||||||||||||||||||||||||
📚 Live Citation Mapping | Partial. Search and Deep Research cite sources in the thread; it is not a dedicated citation-map UI like Perplexity. | No. Not a citation-mapping search product. | No. Not a cited-search product. | Partial. Research and web search can cite sources; there is no dedicated citation-map UI. | No. Not a cited-search product. | No. Not a cited-search product. | Partial. Plans mention keeping citations consistent. That is style, not live academic search. | Yes | Yes, requires free plan. Chat answers include in-line citations back to the notebook sources. | Partial. Search grounding and Deep Research cite links in answers; not a citation-map canvas. | No. Not a cited-search product. | No. Not a cited-search product. | No. Not a cited-search product. | No. Not a cited-search product. | No. Not a cited-search product. | Partial. Bing-backed answers cite web links; there is no academic citation-map UI. | No. Not a cited-search product. | No. Not a cited-search product. | No. Not a cited-search product. | No. Not a cited-search product. | No. Not a cited-search product. | No. Citations are to your meeting, not the academic web. | No. Not a cited-search product. | No. Not a cited-search product. | No. Not a cited-search product. | No. Not a cited-search product. | No. Not a cited-search product. | No. Citations are to your meetings. | No. Not a cited-search product. | No. Not a cited-search product. | Partial. Grounding with Search returns citations. RAG returns source chunks. | No. Not a cited-search product. | Partial. Comet uses Perplexity’s answer engine, which cites sources on perplexity.ai; Comet help does not separately document citation UI in the sidecar. | No. Not a cited-search product. | Partial. Web search shows inline reference links. | Partial. Knowledge Base RAG returns source chunks. Not Perplexity-style inline web citations. | No. Citations are to your meetings. | No. Not a cited-search product. | No. Not a cited-search product. | No. Not a cited-search product. | Unknown. Grounding citations on Azure were not checked. | No. Built-in web search is not a cited academic mapper. | Unknown. A citation generator is part of the product in market; it was not verified this round. | No. Not a cited-search product. | No. Not a cited-search product. | Partial. Assistant highlights citation spans in retrieved chunks. | No. Not a cited-search product. | No. Not a cited-search product. | No. Not a cited-search product. | No. Not a cited-search product. | Unknown. Search heritage, but live citation mapping was not on the Workspace 6.0 homepage. | Partial. Essay FAQ: includes citations and sources with generated information. Not a live academic mapper like Perplexity. | No. Not a cited-search product. | No. Not a cited-search product. | No. Not a search product. | No. Optional web search while generating, not live citation mapping. | No. Not a search product. | No. No live product at tome.app. | ||||||||||||||||||||||||||||||||||||||||||||||||||
| 👨💻 Developer | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
🛠️ Function Calling | Yes. Cursor connects to MCP servers that expose tools the model can execute. Agents call those functions from chat and Plan Mode, with user approval before each tool run by default. | No. Auto populated | Yes, requires pro plan | Yes, requires pro plan | No | Yes, requires pro plan. Built-in tools plus MCP and the Agent SDK invoke structured tool calls. | No | No. No tool-calling interface. | No. No tool-calling interface. | Partial. smolagents and TGI can tool-call. That is libraries/endpoints, not a Hub GPT-store. | No. No tool-calling interface. | No | No. Auto populated | Yes, requires pro plan | No. No tool-calling interface. | No. You POST an image job. No tool-calling chat. | Yes, requires plus plan. MCP server so external agents can read/act on designs. Agent skills/connectors. | No. Rewrite/proofread agents, not a developer tool-calling API. | No. No tool-calling interface. | Yes, requires pro plan | No. No tool-calling interface. | No | No. No tool-calling interface. | Partial. The secure cloud computer can produce files (CSV, JSON, code). That is not a developer tool-calling API for third-party functions. | Yes. Tool calling drives the agentic editing loop. | Yes, requires pro plan | No | No. You POST a generation job and poll. No tool-calling chat. | Yes, requires free plan. Agents, Chatbots, and MCP expose tools. Multi-step Zaps are Professional. | No. No tool-calling interface. | No. Auto populated | No. Auto populated | No. Returns a transcript, not tool calls. | Yes, requires free plan. Agent takes actions (write code, use services, publish) rather than chat-only replies. | No. No tool-calling interface. | No. No tool-calling interface. | No. You submit generate jobs. There is no tool-calling chat interface. | No | Yes, requires free plan. The harness is tool/skill calling: inbox, calendar, flights, code, cron. | No. No tool-calling interface. | Yes, requires free plan. Tool calling is the framework. Human approval for sensitive actions on LangSmith. | No. No tool-calling interface. | Yes, requires free plan. Agents use tools, MCP, shell, and subagents rather than a chat-only reply. | No. No tool-calling interface. | Yes, requires free plan. Agents with 500+ tool superpowers and MCP. | No. You POST generate/edit jobs. No tool-calling chat. | No. Auto populated | Partial. MCP server is listed. That is tool access for assistants, not a public function-calling chat API. | No. No tool-calling interface. | No. No tool-calling interface. You submit a generate-video request. | Yes, requires free plan. Agents run shell/tools and can connect Slack, Linear, and GitHub. | Yes. The xAI API supports function calling: you define custom tools and Grok returns tool-call requests for your app to execute, alongside built-in tools such as web search, X search, and code interpreter. | Partial. MCP exposes Photoroom tools to agents. Not a chat function-calling API. | No. No tool-calling interface. | Yes, requires free plan. Agents take actions (implement, test, hooks, MCP tools), not chat-only replies. | Unknown. Depends on the language model you host. Not a Replicate-wide tools API. | Unknown. AskFred is Q&A. A developer function-calling API was not separately confirmed. | Yes, requires free plan. Edits files, runs linters/tests, and commits. That is tool use, not chat-only. | Yes, requires free plan. Linear Agent plus MCP: agents draft PRs and take issue actions. | No. No tool-calling interface. | Yes, requires plus plan. Function calling in Gemini and agent tool use. | No. Not a developer function-calling API. | Yes, requires free plan. The assistant is documented taking multi-step actions (shop, email, fill forms) rather than only answering. | No. No tool-calling interface. | Yes, requires free plan. AI agent nodes call tools/integrations. Structured I/O is documented. | Yes, requires pro plan. AI Extensions invoke app actions and tools. | Yes, requires plus plan. Bedrock Agents use tool/action groups for function calling. | No. Not a developer function-calling API. | Yes | Yes, requires free plan. Agent SDK tool loops; chat completions can carry tools depending on the model. | Yes, requires plus plan. AI Teammates take approved actions with audit trails and spend limits. | Yes, requires plus plan. Agents loop with coding tools, linters/SAST, and MCP rather than chat-only replies. | Yes, requires plus plan. Functions, tools, and parallel tool calling are listed on Azure OpenAI chat models. | Yes, requires free plan. Tool use on all hosted chat models: local functions, remote MCP, or Groq built-in tools. | No | No. Not a developer tools API. | No. Auto populated | Unknown. Agent tool use was not confirmed. | Unknown. Tool use on hosted models was not confirmed on the homepage. | No. Not an LLM with tools. Assistant may delegate to your stack. | Partial. Modules are tools. An LLM function-calling product was not confirmed. | Yes, requires pro plan. Agents use MCP tools (git, tests, linters, Jira, APIs). | Partial. Voice Agent can invoke configured tools during a call. | Partial. Voice Agent can call tools via LLM Gateway. | Unknown. Claw/agents are named. Tool-calling semantics were not documented. | No | Yes, requires pro plan | No. Not a developer tools API. | Unknown. Agent tool use was not confirmed this round. | Unknown. Tool use in AI features was not confirmed. | No. No tool-calling interface. | No. Not an LLM with tools. | No. Not a developer tools API. | No | No. Not an LLM with tools. | No | No | Yes | No | No | No | No | No | No. No live product at tome.app. | No | No. Auto populated | No | Yes. AI booking and workflow triggers |
⚡ Streaming Responses | Yes. iOS follows agent work live and watches the chat stream as the agent codes. The TypeScript SDK iterates streams; ACP clients handle streaming session updates. | No. Auto populated | Yes | Yes | Yes | Yes, requires pro plan. The CLI and IDE surfaces stream tokens as the agent works. | Yes | No. Generations are returned complete, not streamed token by token. | No. Generations are returned complete, not streamed token by token. | Yes, requires plus plan. Text Generation Inference and Inference Endpoints stream generations. | No. Generations are returned complete, not streamed token by token. | Yes | Yes. Auto populated | Yes | No. Generations are returned complete, not streamed token by token. | No. REST v2beta supports polling for async jobs, not token streaming. | Unknown. Agent streaming was not specified. | Unknown. Token streaming is not documented as an API contract. | No. Generations are returned complete, not streamed token by token. | Yes | No. Generations are returned complete, not streamed token by token. | Yes | No. Generations are returned complete, not streamed token by token. | Unknown. Chat streams in the UI, but token streaming is not documented as a product guarantee. | Yes. Responses stream into the editor as they are produced. | Yes | Yes | No. Async poll for the finished image, not token streaming. | Unknown. Zap runs are jobs. Token streaming of an agent reply was not specified. | No. Generations are returned complete, not streamed token by token. | Yes. Auto populated | No. Auto populated | No. Typical APIs return the finished transcript (json/text). Not token streaming of a chat. | Unknown. Token streaming is not documented as an API contract. | No. Generations are returned complete, not streamed token by token. | No. Generations are returned complete, not streamed token by token. | No. Clips are returned when generation finishes. Failed jobs refund credits rather than streaming tokens. | Yes | Unknown. Realtime voice wiring is mentioned in testimonials. Token streaming of the desktop chat was not specified. | No. Generations are returned complete, not streamed token by token. | Yes, requires free plan. Streaming tokens is standard in the Python/JS APIs. | No. Generations are returned complete, not streamed token by token. | Unknown. Token streaming in the agent UI is not documented as an API guarantee. | No. Generations are returned complete, not streamed token by token. | Unknown. Not specified on the homepage. | No. API returns image URLs. Not token streaming. | Yes. Auto populated | Yes, requires free plan. Live captions stream as people talk. | No. Generations are returned complete, not streamed token by token. | No. Clips are returned when generation finishes, not token-streamed. | Yes, requires free plan. A terminal streams command and agent output as it runs. | Yes. The xAI API streams Grok text output over Server-Sent Events. Set stream: true on chat completions; all text-capable Grok models support it, including grok-4.6. | No. Edits return images, not token streams. | No. Generations are returned complete, not streamed token by token. | Unknown. Token streaming is not documented as an API contract. | Yes, requires plus plan. Docs include streaming output for compatible models. | Yes, requires free plan. Live notes during the call. | Unknown. Streaming depends on the connected LLM. Not documented as an Aider API guarantee. | Unknown. Agent token streaming was not specified. | No. Generations are returned complete, not streamed token by token. | Yes, requires plus plan. Streaming generateContent responses. | No. Editor returns assets, not token streams. | Unknown. Token streaming in the sidecar is not called out in Comet help. | No. Generations are returned complete, not streamed token by token. | Yes, requires free plan. Pricing table lists streaming and bulk operations. | Yes, requires pro plan. AI responses stream in the launcher. | Yes, requires plus plan. InvokeModelWithResponseStream and Converse stream tokens. | Yes, requires free plan. Live notes during the meeting. | Yes | Yes, requires free plan. Quickstart: the API supports streaming. | Unknown. Not specified. | Unknown. Token streaming is not documented as an API contract. | Yes, requires plus plan. Chat Completions and Responses APIs stream tokens; the service is documented as processing prompts via those APIs. | Yes, requires free plan. Chat completions support streaming. Groq’s pitch is low-latency tokens. | No | Unknown. Not confirmed. | No. Auto populated | Partial. Agent responses may stream in the UI. | Yes, requires plus plan. Serverless inference streams tokens. Research kernels (FlashAttention) are the latency pitch. | Partial. Assistant may stream answers. Index upserts are batch/API. | No. Scenario runs are jobs, not token streams. | Yes, requires plus plan. Inline completions stream into the editor as you type. | Yes, requires plus plan. Realtime streaming STT and TTS. | Yes, requires plus plan. Realtime STT streams partial transcripts over WebSocket. | Unknown. Not confirmed on the public homepage this round. | Yes | Yes | Unknown. Not specified. | Partial. Realtime streaming avatars for interactive agents. | Partial. AI writing suggestions stream inline. | No. Generations are returned complete, not streamed token by token. | Partial. Generation may stream audio output in the web app. | Unknown. Not specified. | Yes | Unknown. Realtime processing was not observable. | No | Yes | Yes | Yes | No | No | Yes | No | No. No live product at tome.app. | Yes | Yes. Auto populated | Yes | Yes |
🎛️ Fine-Tuning | No. Public APIs run hosted agent workflows. Docs do not offer a customer fine-tune SKU for foundation weights; you pick Cursor and third-party hosted models. | No. Auto populated | No | Yes, requires pro plan | Yes | No. Claude Code runs Anthropic-hosted models; no custom-weight training is offered. | No | No. No custom model training. | No. No custom model training. | Yes, requires free plan. PEFT, TRL, and Hub model training workflows are first-party libraries. | No. No custom model training. | No | No. Auto populated | No | Partial. Custom style and character training on paid tiers. | Yes, requires free plan. SD 3.5 is marketed as customizable (fine-tunes/LoRA). Brand Studio adds Brand ID models on Enterprise. | No. Uses off-the-shelf and fine-tuned models they host. You do not train Figma. | No. Brand tones/style guides are settings, not custom-weight training. | No. No custom model training. | No | No. No custom model training. | No | Partial. Custom avatars and cloned voices, not model fine-tuning. | No. You ground on sources. There is no custom-weight training product. | No. No custom model training. | No | Yes | Yes, requires plus plan. Open-weight [dev]/[klein] plus paid self-host licenses list fine-tuning and LoRA rights. API OpenAPI also lists fine-tune management. | No. Zapier does not train foundation weights. | Partial. Custom Models can be trained on your own approved assets. | No. Auto populated | No. Auto populated | Partial. Weights are downloadable so the community fine-tunes. OpenAI does not sell a Whisper fine-tune SKU on the GitHub page. | No. Intelligent model routing picks hosted models. No custom-weight training product. | No. No custom model training. | No. No custom model training. | No. No custom-weight training product is documented. | No | No. You bring model subscriptions. OpenClaw does not train foundation weights. | No. No custom model training. | No. You pick models. LangChain does not train weights. | No. No custom model training. | No. You pick Gemini (and on Ultra, third-party) models. No custom-weight training product. | No. No custom model training. | No. Uses GPT/Claude/Gemini. ClickUp does not train those weights. | Yes, requires enterprise plan. Custom Models: train on-brand image models from approved assets. Enterprise/sales for private workflows. | No. Auto populated | No. You do not train Otter’s speech model. | No. No custom model training. | No. No custom-weight training product for Veo. | No. You bring models. Warp does not train custom weights. Self-improvement loops edit factory config, not a foundation model. | No | No. You do not train Photoroom’s models. | Partial. Custom styles can be trained from your own reference images. Not full model fine-tuning. | No. You pick models (Auto, Claude, GPT, open-weight). No custom-weight training product. | Yes, requires plus plan. Trainings API: fine-tune image models (e.g. FLUX LoRA) and push the result. | No. You do not train Fireflies’ models. | No. You pick a model. Aider does not train weights. | No. Linear does not train foundation weights. | No. No custom model training. | Yes, requires plus plan. Supervised fine-tuning and tuning jobs in Vertex AI. | No. You do not train Express. | No. Comet runs hosted Perplexity/partner models; there is no custom-weight training product. | Yes. Train custom models on your own image datasets. | No. You connect models. n8n does not train weights. | No. Routes to third-party models; does not fine-tune. | Yes, requires plus plan. Custom model fine-tuning is a listed Bedrock capability. | No. You do not train Fathom. | Yes. Open weights | No. You consume providers. OpenRouter does not train your weights. | No. Asana does not train foundation weights. | No. Learnings adapt review style. No custom-weight training product. | Yes, requires plus plan. Customers can upload training data to create a fine-tuned model exclusive to their resource. | No. You pick hosted models. Groq is inference, not a fine-tune product. | No | No. You do not train QuillBot. | No. Auto populated | No. Not a model fine-tuning product. | Yes, requires plus plan. Fine-tuning is a listed product next to inference and GPU cloud. | No. Does not train foundation models. | No. Make does not train foundation weights. | No. You connect org context and optionally your LLM. No Tabnine custom-weight training SKU is listed. | Unknown. Custom STT models were not confirmed on homepage. | Unknown. Custom STT was not confirmed on homepage. | No. No fine-tune product is advertised on genspark.ai. | No | Yes, requires pro plan | No. You do not train Wordtune. | No. Voice clone and avatars, not FM fine-tuning. | No. Does not fine-tune models. | Partial. Custom style training from your own images. | Partial. Community LoRA mentioned; not a first-party fine-tune SKU. | No. Powered by Anthropic per FAQ. You do not train it. | No | No. Not a fine-tuning product. | Yes | No | Yes, requires enterprise plan | No | No | No | No | No | No. No live product at tome.app. | No | No. Auto populated | No | No |
📐 Embeddings API | No. Cursor APIs are agent, admin, and analytics endpoints, not a vector embeddings API for converting text into embeddings in your own app. | No. Auto populated | No | Yes, requires pro plan | No | No. Not an embeddings product. | No | No. No embeddings endpoint. | No. No embeddings endpoint. | Yes, requires plus plan. Inference Providers/Endpoints serve embedding models from the Hub. | No. No embeddings endpoint. | No. Gamma's public API generates presentations, documents, websites, and social posts via /v1.0/generations. The documented API options do not include an embeddings or vector-search endpoint. | No. Auto populated | No | No. No embeddings endpoint. | No. Not an embeddings API. | No. Not an embeddings API. | No. Not an embeddings API. | No. No embeddings endpoint. | Yes, requires pro plan | No. No embeddings endpoint. | No | No. No embeddings endpoint. | No. Not an embeddings API. Retrieval stays inside the notebook. | No. No embeddings endpoint. | Yes, requires pro plan | No | No. Not an embeddings API. | No. Not an embeddings API. | No. No embeddings endpoint. | No. Auto populated | No. Auto populated | No. Not an embeddings API. | No. Not an embeddings API. | No. No embeddings endpoint. | No. No embeddings endpoint. | No. Not an embeddings API. | No | No. Not an embeddings API product. | No. No embeddings endpoint. | Partial. Embedding integrations exist in the OSS stack. LangChain is not an embeddings vendor. | No. No embeddings endpoint. | No. Not an embeddings API. | No. No embeddings endpoint. | No. Not an embeddings vendor. | No. Not an embeddings API. | No. Auto populated | No. Not an embeddings API. | No. No embeddings endpoint. | No. Not an embeddings API. | No. Codebase indexing is not a public embeddings API. | No | No. Not an embeddings API. | No. No embeddings endpoint. | No. Not an embeddings API. | Unknown. You can run embedding models if published. Not a first-party embeddings SKU. | No. Not an embeddings API product. | No. Repo map is not a public embeddings API. | No. Not an embeddings API. | No. No embeddings endpoint. | Yes, requires plus plan. Text embeddings and Vector Search. | No. Not an embeddings API. | No. Not an embeddings API product. | No. No embeddings endpoint. | Partial. RAG setups are documented. That uses whatever embedding API you wire, not an n8n embeddings product. | No. Not an embeddings product. | Yes, requires plus plan. Titan Embeddings and Knowledge Base vector indexes. | No. Not an embeddings API. | Yes | Yes, requires plus plan. Client SDK docs cover embeddings alongside streaming. | No. Not an embeddings API. | No. Not an embeddings API. | Yes, requires plus plan. Embedding models convert text to vectors in the Azure OpenAI catalog. | Unknown. An embeddings endpoint was not on the quickstart page checked. | No | No. Not an embeddings API. | No. Auto populated | No. Not an embeddings product. | Unknown. An embeddings endpoint was not confirmed on the homepage. | Yes, requires plus plan. Hosted embedding and rerank models are a core SKU. | No. Not an embeddings API. | No. Context Engine is not a public embeddings API. | No. Speech APIs, not embeddings. | No. Not an embeddings product. | Unknown. Not confirmed on the public homepage this round. | No | Yes, requires pro plan | No. Not an embeddings API. | No. Not an embeddings product. | No. Not an embeddings product. | No. No embeddings endpoint. | No. Not an embeddings product. | No. Not an embeddings API. | No | No. Not embeddings. | No | No | Yes. Industry-leading embed models | No | No | No | No | No | No. No live product at tome.app. | No | No. Auto populated | Yes. Via Hugging Face Inference API | No |
📋 Structured Output (JSON) | Unknown. ACP and SDKs exchange JSON-RPC/tool payloads, but no first-party JSON-schema guarantee for model replies was documented on the API overview. | No. Auto populated | Yes | Yes | No | Unknown. JSON-schema guarantees are not documented as a Claude Code product switch. | No | No. Not a language model, so this does not apply. | No. Not a language model, so this does not apply. | Unknown. Depends on the served model. Not a Hub-wide JSON-schema guarantee. | No. Not a language model, so this does not apply. | No | No. Auto populated | Yes | No. Not a language model, so this does not apply. | No. Output is an image, not JSON schema. | Partial. Variables, Code Connect, MCP. Not arbitrary JSON chat. | No. Output is revised prose, not JSON schema. | No. Not a language model, so this does not apply. | Yes | No. Not a language model, so this does not apply. | No | No. Not a language model, so this does not apply. | Partial. Cloud-computer outputs can include CSV and JSON. Everyday chat is cited prose, not a JSON-schema API. | No. Not a language model, so this does not apply. | Yes | No | No. Output is an image (jpeg/png/webp), not JSON schema. | Yes, requires pro plan. AI fields, webhooks, and structured Zap data. Professional unlocks the full builder. | No. Not a language model, so this does not apply. | No. Auto populated | No. Auto populated | Partial. verbose_json with word/segment timestamps. Not arbitrary JSON schema. | Partial. Agent 4 can flesh out requirements before a build. Not a JSON-schema chat API. | No. Not a language model, so this does not apply. | No. Not a language model, so this does not apply. | No. Output is video or stills, not JSON schema. | No | Unknown. Skills return actions. A JSON-schema guarantee was not on the homepage. | No. Not a language model, so this does not apply. | Yes, requires free plan. Structured output parsers / tool args are a core pattern. | No. Not a language model, so this does not apply. | Unknown. Artifacts exist in the IDE; JSON-schema guarantees were not documented. | No. Not a language model, so this does not apply. | Partial. Tasks, custom fields, dashboards. Not JSON-schema chat. | No. Output is an image. JSON prompting is an input style, not JSON-schema output. | No. Auto populated | Partial. Summaries, action items, and speakers. Not arbitrary JSON schema. | No. Not a language model, so this does not apply. | No. Output is video, not JSON schema. | Partial. Evals/benchmarks score spec adherence. Not a JSON-schema chat API. | No | No. Returns images/video, not JSON answers. | No. Not a language model, so this does not apply. | Partial. Specs and property-based tests are structured artifacts. Not a JSON-schema chat API. | Partial. Cog/input schemas. JSON-mode depends on the model. | Partial. Summaries, action items, conversation intelligence. Not arbitrary JSON schema. | No. Output is code diffs/commits, not JSON schema. | Partial. Issues, labels, and API objects. Not arbitrary JSON-schema chat. | No. Not a language model, so this does not apply. | Yes, requires plus plan. JSON schema / response schema in Gemini API. | No. Returns designs, not JSON answers. | Unknown. JSON-schema guarantees are not documented for the Comet assistant. | No. Not a language model, so this does not apply. | Yes, requires free plan. Enforce structured inputs/outputs around AI steps. | Unknown. JSON mode depends on the selected model. | Partial. Tool schemas and JSON mode depend on the model. Guardrails add structure. | Partial. Summaries and action items. Not JSON schema. | Yes | Unknown. JSON-schema guarantees depend on the routed model; not asserted here. | Yes, requires free plan. Work Graph: who, what, when, goal, dependency. | Partial. Pre-merge checks and reports are structured. Not a JSON-schema chat API. | Yes, requires plus plan. Structured outputs are listed on gpt-chat-latest and other Azure OpenAI chat models. | Yes, requires free plan. JSON mode is listed for the hosted chat models. | No | No. Returns rewritten text, not JSON schema. | No. Auto populated | No. Outputs video projects, not JSON schemas. | Unknown. JSON mode was not confirmed on the homepage. | No. Returns vectors and metadata, not LLM JSON mode. | Yes, requires free plan. Mapped JSON between modules. | Unknown. JSON-schema guarantees were not documented on the pricing page. | Partial. Structured transcript formats and diarization. | Partial. Speech Understanding returns structured entities and summaries. | Partial. Live dashboards from English queries are structured views. JSON-schema API was not listed. | No | No | No. Returns sentences, not JSON schema. | No. Outputs video/audio, not JSON schemas. | No. Outputs mail and docs, not JSON schemas. | No. Not a language model, so this does not apply. | No. Outputs audio files, not JSON schemas. | Partial. Outline-first then Smart Slides. Not JSON schema. | No | No. Outputs enhanced audio, not JSON. | No | No | Yes | No | No | No | No | No | No. No live product at tome.app. | No | No. Auto populated | No | No |
🧩 Reasoning / Thinking Mode | Yes. The model catalog lists Thinking variants and GPT reasoning-effort options (for example gpt-5-high). The SDK exposes thinking messages and a reasoning-effort parameter. | No. Auto populated | Paid Only. Auto populated | Yes. Auto populated | No | Yes, requires pro plan. Runs Claude models with extended thinking for complex work. | No. Auto populated | No. Not a reasoning model. | No. Not a reasoning model. | Partial. You can host/run reasoning models. The Hub has no single Deep Think switch. | No. Not a reasoning model. | No | No. Auto populated | Yes. Auto populated | No. Not a reasoning model. | No. Not an LLM reasoning mode. | No. Figma agent is design automation, not Deep Think. | No. Not an LLM Deep Think product. Humanizer/AI Detector are writing agents. | No. Not a reasoning model. | Yes. DeepSeek R1 reasoning model | No. Not a reasoning model. | No | No. Not a reasoning model. | Unknown. Overlapping columns disagreed (yes vs partial); left unverified rather than copying a no. | Yes. Available through the frontier models it can be pointed at. | Yes. Auto populated | No | No. Not an LLM reasoning mode. | No. Copilot helps build Zaps. That is not a Deep Think reasoning mode. | No. Not a reasoning model. | No. Auto populated | No. Auto populated | No. Not an LLM reasoning mode. | Unknown. Overlapping columns disagreed (yes vs partial); left unverified rather than copying a no. | No. Not a reasoning model. | No. Not a reasoning model. | No. Not a chain-of-thought LLM. VIDEO 3.0 is a generator with multi-shot planning in the prompt. | Yes. Auto populated | Unknown. Depends on the model (GPT-5.5, Claude, etc.). No OpenClaw Deep Think toggle. | No. Not a reasoning model. | Partial. Deep Agents for long-running work. That is agent loops, not a user Deep Think toggle. | Partial. Inherits reasoning from the underlying frontier models it routes to. | Yes, requires free plan. Core agent models include Gemini 3.1 Pro / 3.5 Flash. Ultra adds third-party models. | No. Not a reasoning model. | Partial. Brain² picks frontier models. Not a user Deep Think toggle. | No. Not an LLM reasoning mode. Magic Prompt rewrites prompts. | No. Auto populated | No. Not an LLM reasoning-mode product. | No. Not a reasoning model. | No. Not a chain-of-thought LLM. Conversational video reasoning is Gemini Omni. | Unknown. Overlapping columns disagreed (yes vs partial); left unverified rather than copying a no. | Yes. Auto populated | No. Not an LLM reasoning mode. | No. Not a reasoning model. | Unknown. Overlapping columns disagreed (yes vs partial); left unverified rather than copying a no. | Unknown. Hosts reasoning models (e.g. Claude). No Replicate Deep Think switch. | No. AskFred is retrieval over meetings, not a Deep Think mode. | Unknown. Overlapping columns disagreed (yes vs partial); left unverified rather than copying a no. | No. Agent runs issues. Not a user Deep Think toggle. | No. Not a reasoning model. | Partial. Thinking/reasoning models in Gemini 2.x family. Platform-level, not a single toggle. | No. Not an LLM reasoning mode. | Partial. Max Assistant on Comet is documented as a slower, routed reasoning mode for Max subscribers, not a free Deep Think toggle. | No. Not a reasoning model. | Yes, requires plus plan. AI Assistant (preview) plus inspectable agent reasoning on the canvas. | Unknown. Overlapping columns disagreed (yes vs partial); left unverified rather than copying a no. | Partial. Reasoning models (e.g. Claude extended thinking) can be hosted. No Bedrock-branded Deep Think toggle. | No. Ask Fathom is retrieval, not Deep Think. | Partial. Llama 3.3 has some CoT capability | Unknown. Overlapping columns disagreed (yes vs partial); left unverified rather than copying a no. | No. Teammates run workflows. Not a Deep Think toggle. | Yes, requires plus plan. Reviews reason over Codegraph/blast radius. Not a user-toggled Deep Think switch. | Yes, requires plus plan. o-series and later chat models list reasoning; some models expose a configurable reasoning effort. | Unknown. Overlapping columns disagreed (yes vs partial); left unverified rather than copying a no. | No | No. Paraphraser, not Deep Think. | No. Auto populated | No. Not a reasoning LLM toggle. | Unknown. You can host reasoning open models. No Together Deep Think toggle. | No. Not a reasoning LLM product. | No. Not an LLM reasoning-mode product. | Yes, requires plus plan. Chat can use leading LLMs (Anthropic, OpenAI, Google, Meta, Mistral, and others). | No. Not a text reasoning model product. | No. Not a reasoning LLM. | Unknown. Not confirmed on the public homepage this round. | No | No | No. Rewrite suggestions, not Deep Think. | No. Not a reasoning LLM product. | Unknown. Reasoning models in AI features were not confirmed. | No. Not a reasoning model. | No. Not a reasoning LLM. | No. Outline-then-slides, not Deep Think. | No | No. Not an LLM. | No | Yes. Access to o1, R1 via Poe | Partial. Auto populated | No | No | No | No | No | No. No live product at tome.app. | No. Auto populated | No. Auto populated | Yes. Auto populated | No |
🏠 Local Execution | No. Closed weights; runs only on OpenAI infrastructure. | No. The CLI runs on your machine; the model stays on Anthropic (or a third-party provider) infrastructure. | Yes, requires free plan. Transformers/Diffusers/llama.cpp-style downloads. Transformers.js runs in the browser. | No. Closed weights; runs only on Anthropic infrastructure and partner clouds. | Yes, requires free plan. Weights and inference code are on Hugging Face/GitHub. SD 3.5 Medium is documented at ~9.9 GB VRAM. | No. Cloud editor. Make-on-local-codebase is still a Figma feature, not self-hosted Figma. | No. Cloud service. Enterprise BYOK is key management, not a local model. | Yes. Open weights can be self-hosted; the largest variants need substantial GPU memory. | No. The new “cloud computer” is hosted by Google, not a local runtime. | Partial. Gemini itself is closed, but the smaller Gemma models are downloadable and run locally. | Yes, requires free plan. Official GitHub inference repo runs open-weight FLUX.1 [schnell] and [dev] on your hardware. | No. Cloud SaaS. Not self-hosted. | Yes, requires free plan. Official GitHub repo runs on your hardware. | No. Code runs on Replit’s cloud, not as a local runtime you download. | No. Hosted by Kuaishou. Weights are not for local download. | No. Microsoft Copilot runs in the cloud, not as a local model you download. | Yes, requires free plan. Open-source agent that runs on your machines. Cloud models are optional BYOK. | Yes, requires free plan. Open-source frameworks run locally. LangSmith cloud is optional. | Yes, requires free plan. Agents run locally on the machine; models remain Google-hosted Gemini (and partners). | No. Cloud SaaS. | Partial. Developer overview lists open weights for Ideogram 4.0. Hosted API remains the documented production path. | No. Cloud SaaS. Desktop recorder still sends audio to Otter. | No. Hosted by Google. Weights are not for local download. | Yes, requires free plan. The terminal and Agent CLI run locally. Cloud agents are optional. Homepage also says the terminal is open source. | No. Cloud editor + API. | Yes, requires free plan. IDE and Crew run on your machine; web sandboxes are cloud. Models are hosted. | Partial. Cog is open source so you can run locally. Hosted Replicate is cloud GPUs. | No. Cloud SaaS. | Yes, requires free plan. The program runs on your machine. Models can be local or cloud APIs you configure. | No. Cloud SaaS. Agent coding sessions run in Linear’s environment. | No. Cloud GCP inference. Distinct from Antigravity local dev. | No. Cloud editor. | No. The browser runs locally; model inference stays on Perplexity’s cloud. | Yes, requires free plan. Community Edition self-hosts (Docker). Cloud is optional. | Yes, requires pro plan. Ollama local models run on-device. | No. Managed AWS inference. Distinct from Kiro local dev. | No. Cloud SaaS. | Yes. Weights are downloadable and run locally via llama.cpp, Ollama and similar runtimes. | No. Routes to cloud providers. Not a local runtime. | No. Cloud SaaS. | Partial. CLI/IDE review local diffs; Enterprise offers self-hosting. Models are otherwise hosted. | No. Cloud-hosted in your Azure region. | No. Runs on Groq LPUs in their cloud, not on your GPU. | No. Cloud web app. | No. Cloud-rendered video editor. | No. Cloud GPUs / serverless. Not on your laptop. | No. Managed cloud vector DB. | No. Cloud SaaS, unlike self-hosted n8n. | Yes, requires plus plan. On-prem/air-gapped deploy plus local IDE/CLI. Unlimited usage when using your own LLM on-prem. | Yes, requires enterprise plan. Self-hosted deployment option alongside cloud. | No. Cloud-only STT. Distinct from self-hosted Whisper. | No. Hosted web workspace, not a local model runtime. | Partial. Several models ship open weights; the largest ones are API-only. | No. Cloud + extension. | No. Cloud-rendered avatar video. | No. Cloud-synced mail and docs. | Partial. Community ONNX CPU inference mentioned; hosted API is cloud. | No. Cloud SaaS. | No. Cloud web tool. | Partial. Command models publish open weights for research; commercial use runs through the API or a private deployment. | No. Nothing to run. URL 404. | Yes. Serves open models that can equally be downloaded and run locally. | ||||||||||||||||||||||||||||||||||||||||||||||
🔓 Open Weights | No. Closed weights; GPT models are not downloadable. | No. Claude model weights are not published. | Yes, requires free plan. The Hub is the public weight catalog. Many (not all) models are open. | No. Closed weights; Claude models are not downloadable. | Yes, requires free plan. SD 3.5 Large/Turbo/Medium are an open release under the Stability AI Community License. | No. Closed product. | No. Closed product, not an open-weight model release. | Yes. DeepSeek R1 | No. Closed Gemini models. | Partial. Gemini itself is closed; Gemma weights are downloadable. | Yes, requires free plan. FLUX.1 [schnell] is Apache-2.0; [dev] is a non-commercial license. Hosted Pro/Max stay closed. | No. Closed platform. | Yes, requires free plan. Model weights and inference code are public on GitHub (108k+ stars). | No. Hosted models via routing. Replit is not an open-weight model release. | No. Closed model. | No. Closed weights on Microsoft-hosted models. | Partial. The harness is open source (most-starred GitHub repo claim). You choose closed or open models. | Partial. Framework is OSS (350M+ monthly downloads). You bring any model, including open weights. | No. The harness is Google’s. Open-weight models are not the documented default. | No. Closed product wrapping frontier APIs. | Yes, requires free plan. Ideogram 4.0 is documented as available across app, API, MCP, and open weights. | No. Closed speech models. | No. Closed model. | Partial. You can use open-source models and BYOK. Warp is not itself a model-weight release. | No. Closed models. | Partial. You can select open-weight models (Qwen, DeepSeek, MiniMax, GLM). The Kiro product itself is not an open-weight release except Crew. | Partial. Hosts many open models. The platform and many popular models are mixed open/closed. | No. Closed models. | Partial. Aider itself is open source on GitHub. It can drive local open-weight LLMs; it is not a model release. | No. Closed product. | Partial. Model Garden hosts open models (Gemma, Llama) plus Google proprietary models. | No. Closed Adobe models. | No. Comet is not an open-weights model release. | Partial. Source is on GitHub. You can swap cloud or offline models. n8n is not a foundation-model release. | Partial. Ollama routes to open-weight local models. | Partial. Hosts some open models (e.g. Llama) alongside proprietary FMs. | No. Closed product. | Partial. You can pick open models hosted by providers. OpenRouter is not a weight release. | No. Closed product. | No. The reviewer is a hosted product, not an open-weight model release. | No. Hosts OpenAI proprietary models, not open-weight releases. | Partial. Hosts open models (Llama, gpt-oss, Qwen). Groq’s runtime/hardware is not open weights. | No. Closed product. | No. Proprietary video AI stack. | Partial. Inference cloud for open models. Together’s serving stack is not a weight release. | No. Hosts embedding models; the platform is not an open-weight release. | No. Closed cloud product. | Partial. You may attach Meta/Mistral and other models, including self-hosted. Tabnine is not an open-weight release. | No. Proprietary Flux models. Distinct from Whisper open weights. | No. Proprietary models and APIs. | No. Closed product. No weights download is advertised. | No. Closed product. | No. Uses proprietary avatar tech; SD portrait is a feature, not open weights. | No. Proprietary product stack. | Partial. Some open/community weights exist; flagship models are Stability-hosted. | No. Closed; FAQ says powered by Anthropic. | No. Adobe proprietary audio models. | No. Never open weights; URL 404. | ||||||||||||||||||||||||||||||||||||||||||||||||||
🤖 Multi-Agent Orchestration | Partial. Agent / Operator can run multi-step tool use; it is not a user-facing swarm of sub-agents like Claude Agent Teams. | Yes, requires pro plan. A lead agent can spawn sub-agents and merge results; Agent SDK covers custom orchestration. | Partial. smolagents is a first-party agent library. The Hub is not an orchestrator SaaS. | Yes, requires pro plan. Agent Teams | No. Not a multi-agent orchestrator. | No. One Figma agent plus external MCP clients. Not a swarm orchestrator. | Partial. Humanizer and AI Detector are named agents. Superhuman Go is the broader agent layer. | Partial. Deep Research is an agentic researcher. It is not a general multi-agent SDK. | No. Gems are custom bots, not a user-facing multi-agent swarm in the consumer app. | No. Not a multi-agent orchestrator. | Partial. Agents exist as an add-on. Not a multi-agent research swarm. | No. Not an orchestrator. | Yes, requires pro plan. Pro documents 10 parallel agents. Homepage also highlights parallel task execution. | No. Not a multi-agent orchestrator. | No. Consumer Copilot is a single assistant; Copilot Studio agents are a different product. | Partial. Testimonials of two claws chatting. Not a marketed multi-agent orchestrator SKU. | Yes, requires free plan. LangGraph / Deep Agents for multi-step and multi-actor graphs. | Yes, requires free plan. Command center runs multiple local agents in parallel; CLI manages background subagents. | Yes, requires plus plan. Intake, Assign, PM, Live Answers agents and more. | No. Not a multi-agent orchestrator. | No. Not an orchestrator. | No. Not a multi-agent orchestrator. Flow “Agent” is a studio feature, not Veo’s API. | Yes, requires plus plan. Factories run fleets of coding agents across the SDLC on any model or harness. | Yes. Four-agent backend | No. Not an orchestrator. | Yes, requires free plan. Parallel agents implement tasks; hooks fire on IDE events; Crew coordinates tools. | No. You orchestrate in your code. | No. Not an orchestrator. | No. One pair-programmer process, not a multi-agent swarm product. | Partial. Agents as teammates. Not a multi-agent research swarm. | Yes, requires plus plan. Gemini Enterprise Agent Platform supports multi-agent flows. | No. Not an orchestrator. | Partial. Background Assistants (Max) run tasks asynchronously; that is not a general multi-agent SDK. | Yes, requires free plan. Homepage: multi-agent setups and RAG, inspectable on the canvas. | No. Single chat session, not multi-agent. | Yes, requires plus plan. AgentCore supports multi-agent workflows and supervisor patterns. | No. Not an orchestrator. | Partial. Agent SDK handles tool loops and state. Not a multi-agent swarm product. | Yes, requires plus plan. Launch Planner, Workflow Optimizer, Status Reporter, and custom teammates in one graph. | Yes, requires plus plan. Loops with coding agents, Slack agent, and cloud coding agent are documented add-ons/plans. | Partial. GPT-5.6 series lists multi-agent orchestration as preview on Azure. | Partial. Compound/built-in tools are a single-call agent loop, not a multi-agent orchestrator product. | No. Not an orchestrator. | Unknown. Whether multiple agents coordinate was not confirmed. | No. You orchestrate in your app. | No. You orchestrate agents in your app. | Unknown. AI/agent modules were not confirmed on a live homepage this round. | Yes, requires pro plan. Agentic Platform: autonomous agents, CLI, and optional headless/CI agents. | No. Single voice agent per call. | No. Single voice agent per session. | Unknown. Claw is named. Orchestration of many agents was not documented. | No. Not an orchestrator. | No. Single-agent avatars, not multi-agent orchestration. | No. Not a multi-agent platform. | No. Not an agent platform. | No. Not an orchestrator. | No. Not an agent platform. | No. No live product at tome.app. | ||||||||||||||||||||||||||||||||||||||||||||||||||
| 📱 Platforms | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
🌐 Web App | Yes. Cursor Web at cursor.com/agents starts and manages cloud agents from a browser on any device, alongside desktop and iOS. | Yes. Auto populated | Yes. github.com chat | Yes | Yes | Yes, requires pro plan. claude.ai/code runs cloud sessions in the browser with no local install. | Yes | Yes. Browser-based; no install required. | Yes. Browser-based; no install required. | Yes, requires free plan. huggingface.co plus Spaces as hosted ML apps. CPU Basic Spaces are free. | Yes. Browser-based; no install required. | Yes | Yes. Auto populated | Yes | Yes. Browser-based; no install required. | Yes, requires free plan. Brand Studio and Stable Assistant are documented web apps. First 1000 Brand Studio credits are a trial. | Yes, requires free plan. figma.com editor. | Yes, requires free plan. grammarly.com is the web editor and account/plans surface. | Unknown. Primarily a CLI; a hosted web app is not confirmed. | Yes | Yes. Browser-based; no install required. | Yes | Yes. Browser-based; no install required. | Yes, requires free plan. Standalone app at notebooklm.google.com / notebooklm.google. | No. Desktop editor only. | Yes | Yes | Yes, requires free plan. Docs point at a playground plus the dashboard at dashboard.bfl.ai. | Yes, requires free plan. zapier.com editor, Tables, Forms, Agents, Chatbots. | Yes. Browser-based; no install required. | Yes. Auto populated | No. Auto populated | No. No official Whisper web app. Hosted playgrounds are third-party (e.g. Groq). | Yes, requires free plan. replit.com is the IDE, Agent, and hosting dashboard. | Yes. Browser-based; no install required. | Yes. Browser-based; no install required. | Yes, requires free plan. The international studio is a browser app at klingai.com / kling.ai. | Yes | No. Install desktop apps (or one-liner). Not a hosted SaaS web chat. | Yes. Browser-based; no install required. | Yes, requires free plan. langchain.com plus LangSmith web console. | Yes. Browser-based; no install required. | No. Surfaces are desktop, IDE, and CLI. A browser IDE is not the documented product. | Yes. Browser-based; no install required. | Yes, requires free plan. clickup.com. | Yes, requires free plan. ideogram.ai is the consumer web app; the API Dashboard has a playground. | Yes. Auto populated | Yes, requires free plan. otter.ai web app for notes and Otter Chat. | Yes. Browser-based; no install required. | Yes, requires free plan. Used in the browser via Flow, Gemini, Google Vids, and Google AI Studio. | Partial. Cloud agents and session links are browser-visible. The primary product is still the desktop terminal. | Yes | Yes, requires free plan. photoroom.com web editor. | Yes. Browser-based; no install required. | Yes, requires plus plan. Kiro on the web (preview) is for paid Pro+ tiers; sandboxes keep running in the cloud. | Yes, requires free plan. replicate.com playground and dashboard. | Yes, requires free plan. fireflies.ai web app. | No. No Aider web IDE. Optional copy/paste to a vendor web chat. | Yes, requires free plan. linear.app. | Yes. Browser-based; no install required. | Yes, requires plus plan. Google Cloud console and Vertex AI Studio. | Yes, requires free plan. express.adobe.com (sign-in required to enter the editor). | No. You download Comet. The marketing site is not a browser-in-the-browser web app. | Yes. Browser-based; no install required. | Yes, requires free plan. Cloud editor at n8n.io; also a self-hosted web UI. | No. Native macOS app, not a web chat. | Yes, requires plus plan. Bedrock console in the AWS Management Console. | Yes, requires free plan. fathom.video web app. | No | Yes, requires free plan. openrouter.ai chat, models catalog, and credit dashboard. | Yes, requires free plan. asana.com. | Yes, requires free plan. coderabbit.ai is the product UI (PR board, triage, settings) plus Git-host comments. | Yes, requires plus plan. Azure AI Foundry / portal for deploying and testing models. | Yes, requires free plan. Playground on console.groq.com. | Yes | Yes, requires free plan. quillbot.com (reached a Cloudflare challenge, but the host is a web app). | No. Auto populated | Yes, requires free plan. invideo.io browser editor. Freemium. | Yes, requires free plan. together.ai console. | Yes, requires free plan. pinecone.io console for indexes and Assistant. | Yes, requires free plan. make.com visual editor (cloud). | Partial. Admin/analytics are documented. The coding surface is the IDE, not a chat website. | Partial. Console for keys and testing. Primary surface is API. | Partial. Dashboard for API keys and playground. | Yes, requires free plan. genspark.ai is a web workspace. | Yes | Yes | Yes, requires free plan. wordtune.com. | Yes, requires plus plan. Creative Reality Studio in the browser. | Yes, requires pro plan. Superhuman mail and docs in the browser. Paid seats. | Yes. Browser-based; no install required. | Yes, requires plus plan. stableaudio.com web generator. | Yes, requires plus plan. beautiful.ai web app. Trial then paid. | Yes | Yes, requires free plan. podcast.adobe.com web tool (homepage empty but URL resolves). | Yes | Yes | Yes | Yes | Yes. Fliki runs entirely in the browser. Paste a script, idea, or blog URL and generate video in the web app — no desktop install required. | Yes | Yes | No | No. Checked https://tome.app on 2026-09-02; HTTP 404. Not a working web app. | Yes | Yes. Auto populated | Yes | Yes |
📱 iOS App | Yes. Native Cursor for iOS (App Store) directs cloud and local agents, reviews PRs, and streams work. Requires iOS/iPadOS 26+; listed on Start through Enterprise. | No. Auto populated | Yes | Yes | Yes | Yes, requires pro plan. The Claude iOS app has a Code tab for cloud sessions and Remote Control. | Yes | Yes. Native iOS app with Magic Studio features. | Yes. Native iOS app. | No. No official iOS Hub app is documented. | Yes. Native iOS app. | No | No. Auto populated | Yes | Yes. Native iOS app. | No. No official iOS app is documented on the product pages checked. | Partial. Figma Make on mobile is listed. The full Design editor is not the iPhone app. | Unknown. A current iOS app was not confirmed on the homepage/plans pages checked. | No. No native mobile app; the web app runs in a mobile browser. | Yes | Yes. Native iOS app. | Yes | No. No native mobile app; the web app runs in a mobile browser. | Yes, requires free plan. Google AI Pro help lists Gemini Notebook on web, iOS, and Android. | No. No native mobile app; the web app runs in a mobile browser. | Yes | No | No. No official iOS app is documented. | No. Zapier's own Shortcuts article says Zapier does not have an App Store app; iPhone use is via Apple Shortcuts plus Webhooks. | Partial. Available through Adobe Express on iOS rather than a standalone Firefly app. | No. Auto populated | Yes. Auto populated | No. No official iOS app. | Unknown. A current native iOS app was not confirmed on the homepage/pricing pages checked. | No. No native mobile app; the web app runs in a mobile browser. | No. No native mobile app; the web app runs in a mobile browser. | Yes, requires free plan. Official App Store app “Kling AI: AI Image&Video Maker” (id 6738049229). Listing covers text/image-to-video up to 15s and extension. | Yes | No. No iOS app. You talk to it from WhatsApp/Telegram on the phone. | No. No native mobile app; the web app runs in a mobile browser. | No. No iOS app. | No. No native mobile app; the web app runs in a mobile browser. | No. Downloads are macOS, Windows, and Linux. | Yes. Native iOS app. | Yes, requires free plan. Native iOS listed. | Unknown. A native iOS app was not confirmed on the docs pages checked. | Yes. Auto populated | Yes, requires free plan. Official iOS app. | No. No native mobile app; the web app runs in a mobile browser. | Partial. There is no Veo iOS app. DeepMind lists Gemini as a surface; Gemini has iOS. A dedicated Veo mobile client was not documented. | No. Downloads are macOS, Windows, and Linux. | Yes. Via X app | Yes, requires free plan. iOS app (App Store billing). | No. No native mobile app; the web app runs in a mobile browser. | Yes, requires plus plan. Homepage: sign in on mobile with a paid account. Store listing OS split was not opened this round; Google/Apple both implied by “mobile app”. | No. No official iOS app. | Yes, requires free plan. iOS app is listed. | No. Terminal tool, no iOS app. | Yes, requires free plan. Homepage demo includes an iOS app. | Yes. Native iOS app. | No. No consumer iOS app. Distinct from Gemini mobile. | Unknown. Mobile apps exist in market; not re-verified this round. | Yes, requires free plan. Perplexity launched a Comet iOS app; the product page lists iOS. | Yes. Native iOS app. | No. No iOS app is documented. | No. No iOS app for Raycast AI. | No. No consumer iOS app. AWS console is web-only. | Unknown. Not confirmed on the homepage fetch. | No | No. No iOS app is documented. | Unknown. Mobile apps exist in market; not re-verified on the homepage this round. | No. No iOS app is documented. | No. No consumer iOS app. Distinct from Copilot mobile. | No. No iOS app is documented. | Yes. Via WhatsApp, Instagram, Facebook apps | Unknown. Not confirmed this round. | No. Auto populated | Unknown. Mobile app was not checked. | No. No iOS app. | No. No iOS app. | Unknown. Not confirmed this round. | No. No iOS app is documented. | No. No consumer iOS app. | No. No iOS app. | Partial. Workspace 3.0 documents Speakly as an iOS companion for voice-to-text dictation, not a full Genspark Workspace app on iPhone. | No | No | Yes, requires free plan. Homepage cites a 97% App Store rating. | Partial. Studio available on mobile per product pages. | Partial. Mobile apps are advertised for mail. | No. No native mobile app; the web app runs in a mobile browser. | Unknown. Mobile app was not checked. | Unknown. Not on the homepage. | Yes | Unknown. Mobile app was not checked. | No | Yes | No | No | No | No | Yes | No | No. No live iOS app to install from the official URL. | Yes | No. Auto populated | No | Yes. Lead Connector app |
🤖 Android App | Partial. There is no native Android app; docs say Android is planned. Use cursor.com/agents in Chrome and Install App for a PWA. | No. Auto populated | Yes | Yes | Yes | Yes, requires pro plan. The Claude Android app exposes the same Code tab as iOS. | Yes | Yes. Native Android app with Magic Studio features. | Yes. Native Android app. | No. No official Android Hub app is documented. | Yes. Native Android app. | No | No. Auto populated | Yes | No. No native mobile app; the web app runs in a mobile browser. | No. No official Android app is documented on the product pages checked. | Unknown. Make on mobile is listed without OS. Android was not isolated. | Unknown. A current Android app was not confirmed on the homepage/plans pages checked. | No. No native mobile app; the web app runs in a mobile browser. | Yes | Yes. Native Android app. | Yes | No. No native mobile app; the web app runs in a mobile browser. | Yes, requires free plan. Official Android app help: phones and tablets, Android 10+, Audio Overviews, share-to-notebook from the device. | No. No native mobile app; the web app runs in a mobile browser. | Yes | No | No. No official Android app is documented. | Unknown. No first-party Android app page was checked this round. The Shortcuts article only covers the missing iOS App Store app. | Partial. Available through Adobe Express on Android rather than a standalone Firefly app. | No. Auto populated | Yes. Auto populated | No. No official Android app. | Unknown. A current native Android app was not confirmed on the homepage/pricing pages checked. | No. No native mobile app; the web app runs in a mobile browser. | No. No native mobile app; the web app runs in a mobile browser. | Unknown. Credits policy refers to a Kling AI APP. A Google Play listing was not opened this round. | Yes | No. No Android app. Chat apps are the mobile surface. | No. No native mobile app; the web app runs in a mobile browser. | No. No Android app. | No. No native mobile app; the web app runs in a mobile browser. | No. No Android client is documented. | No. No native mobile app; the web app runs in a mobile browser. | Yes, requires free plan. Native Android listed. | Unknown. A native Android app was not confirmed on the docs pages checked. | Yes. Auto populated | Yes, requires free plan. Official Android app. | No. No native mobile app; the web app runs in a mobile browser. | Partial. Same as iOS: Gemini Android may expose video gen; there is no Veo Android app. | No. No Android client is documented. | Yes. Via X app | Yes, requires free plan. Android app (Play billing). | No. No native mobile app; the web app runs in a mobile browser. | Unknown. A mobile app is documented; Android vs iOS split was not confirmed on a store page. | No. No official Android app. | Yes, requires free plan. Android app is listed. | No. Terminal tool, no Android app. | Unknown. iOS is shown. Android was not confirmed on the homepage. | No. No native mobile app; the web app runs in a mobile browser. | No. No consumer Android app. | Unknown. Not re-verified this round. | Yes, requires free plan. The product page lists Android alongside desktop. | Yes. Native Android app. | No. No Android app is documented. | No. No Android app. | No. No consumer Android app. | Unknown. Not confirmed on the homepage fetch. | No | No. No Android app is documented. | Unknown. Not re-verified on the homepage this round. | No. No Android app is documented. | No. No consumer Android app. | No. No Android app is documented. | Yes. Via WhatsApp, Instagram, Facebook apps | Unknown. Not confirmed this round. | No. Auto populated | Unknown. Mobile app was not checked. | No. No Android app. | No. No Android app. | Unknown. Not confirmed this round. | No. No Android app is documented. | No. No Android app. | No. No Android app. | Partial. The same Workspace 3.0 launch covers Speakly on Android as a dictation companion. It is a voice keyboard, not a full Genspark Workspace mobile client. | No | No | Unknown. “Mobile device” is mentioned. Android was not isolated. | Partial. Studio available on mobile per product pages. | Partial. Mobile apps are advertised for mail. | No. No native mobile app; the web app runs in a mobile browser. | Unknown. Mobile app was not checked. | Unknown. Not on the homepage. | Yes | Unknown. Mobile app was not checked. | No | Yes | No | No | No | No | Yes | No | No. No live Android app from the official URL. | Yes | No. Auto populated | No | Yes. Lead Connector app |
🖥️ Desktop App | Yes. Cursor is a native IDE: .dmg on macOS 12+, .exe on Windows 10+, and apt/yum/AppImage on Linux. Quickstart starts with installing that desktop app. | No. Auto populated | Yes. VS Code, JetBrains, Vim, etc. | Yes | No | Yes, requires pro plan. Standalone macOS and Windows app; docs say a paid subscription is required. | Yes | Yes. Native desktop app for macOS and Windows. | No. Browser only. | No. Web + libraries. PRO mentions SSH/VS Code for Spaces, not a Hub desktop app. | No. Browser only. | No | No. Auto populated | Yes | No. Browser only. | No. Self-host is your own stack (ComfyUI etc.), not a Stability desktop app. | Yes, requires free plan. Official desktop app is a long-standing Figma product (not re-listed on the AI pricing table). | Unknown. A native desktop app was not confirmed on the pages checked this round. | Partial. Runs locally as a CLI rather than as a windowed desktop app. | No | Yes. Native desktop app. | No | No. Browser only. | No. Desktop use is the browser. No native Gemini Notebook desktop client is documented. | Yes. A desktop IDE, forked from VS Code. | Yes, requires free plan. Native Gemini app for Apple Silicon Macs (macOS 15+, Option+Space). Windows gets Gemini in the Google app (Alt+Space). Both are free. | Yes | No. Self-host is your own stack, not a BFL desktop app. | No. Browser product, not a native desktop app. | Yes. Integrated into Photoshop, Illustrator and other desktop apps. | No. Auto populated | No. Auto populated | No. CLI/Python package, not a desktop app. | No. The documented product is the browser IDE, not a native desktop install. | No. Browser only. | No. Browser only. | Unknown. No native desktop client is documented on the marketing site. Web is the desktop path. | No | Yes, requires free plan. Full apps for macOS, Windows, and Linux (AppImage/Debian). They install gateway, chat, setup, and node features. | Yes. Native desktop app. | No. No branded desktop app. | No. Browser only. | Yes, requires free plan. Antigravity 2.0 is a standalone desktop command center; Antigravity IDE is a separate desktop install. | No. Browser only. | Yes, requires free plan. macOS and Windows listed. | No. The documented surfaces are web and API. | No. Auto populated | Yes, requires free plan. Bot-free desktop recorder is advertised. | No. Browser only. | No. No native Veo desktop app. Flow and Gemini are browser/Gemini-app surfaces. | Yes, requires free plan. Native terminal for macOS, Windows 10/11, and Linux (brew, winget, .deb/.rpm/AppImage). | No | No. Web + mobile. No native desktop app advertised. | No. Browser only. | Yes, requires free plan. Downloadable IDE for the OS; Open VSX + VS Code settings import. | No. API + web, not a desktop app. | Yes, requires free plan. Desktop app is listed alongside the Chrome extension. | No. Installs as a Python CLI, not a branded desktop app. | Unknown. A desktop client exists in market; it was not on the pricing page this round. | No. Browser only. | No. Browser console, not a desktop chat app. | Unknown. Not confirmed this round. | Yes, requires free plan. Standalone Mac and Windows browser; setup help covers install and default-browser. | No. Browser only. | No. Docker/self-host or cloud. Not a branded desktop app. | Yes, requires pro plan. Raycast is a macOS desktop launcher with built-in AI. | No. Console is browser-based, not a branded desktop app. | Unknown. Bot-free capture may be desktop; installer was not on the homepage. | No | No. No desktop app is documented. | Unknown. Not confirmed on the homepage. | No. IDE extensions and CLI, not a standalone desktop app. | No. Portal and API. No branded consumer chat desktop app. | No. API + Playground, not a desktop app. | No | Unknown. Not confirmed this round. | No. Auto populated | No. Browser-based editor. | No. No desktop app. | No. Web console only. | No. Cloud editor, not a native desktop app. | No. Installs into existing IDEs plus a CLI, not a standalone Tabnine desktop IDE. | No. No desktop app. | No. No desktop app. | Unknown. GenClipboard/GenTerminal names suggest desktop utilities; installers were not confirmed. | No | No | Partial. “Desktop” is listed as a place it works. A native installer was not specified. | Partial. Desktop studio app is advertised alongside web. | Partial. Native mail clients for Mac and Windows. | No. Browser only. | Partial. DAW plugin runs inside desktop audio software. | No. Browser product. | No | No. Web tool at podcast.adobe.com, not a native desktop app. | No | No | No | No | No | No | No | No | No. No live desktop app from the official URL. | No | No. Auto populated | No | Yes. HighLevel ships native desktop apps for macOS, Windows, and Linux (HighLevel and LeadConnector builds). The app wraps the same workspace as the web product, with native OS notifications for teams that keep it open all day. |
🔵 Chrome Extension | No. Documented clients are the desktop IDE, cursor.com/agents, iOS, Slack, GitHub, Linear, and the API. Agent uses a built-in IDE browser, not an official Chrome extension. | No. Auto populated | Yes. GitHub.com integration | No | No | Yes, requires pro plan. Claude in Chrome is listed as a Code surface for debugging live web apps. | Yes | Yes. Official Chrome extension. | No. No official browser extension. | No. No official Chrome extension. | No. No official browser extension. | No | No. Auto populated | No | No. No official browser extension. | No. No official Chrome extension. | No. Not a Chrome-extension product. | Yes, requires free plan. Grammarly’s in-page assistant is distributed as a browser extension, including Chrome. | No. No official browser extension. | No | Yes. Official Chrome extension. | Yes | No. No official browser extension. | Unknown. Not documented as a Chrome Web Store extension in the help articles checked. | No. No official browser extension. | No | No | No. No official Chrome extension. | Yes, requires free plan. zapier.com/l/install-chrome-extension documents the official Chrome extension to trigger Zaps and show Reply to Chrome results without leaving the tab. | No. No official browser extension. | No. Auto populated | No. Auto populated | No. No official Chrome extension. | No. No official Chrome extension is documented. | No. No official browser extension. | No. No official browser extension. | No. No official Chrome Web Store extension is documented. | Yes | No. No Chrome extension is advertised. | No. No official browser extension. | No. No Chrome extension. | No. No official browser extension. | No. No official Chrome extension is documented. | No. No official browser extension. | Unknown. Not confirmed on the homepage. | No. No official Chrome extension is documented. | No. Auto populated | Yes, requires free plan. Chrome capture extension. | No. No official browser extension. | No. No official Chrome extension. | No. No official Chrome extension. | No | Unknown. Not confirmed this round. | No. No official browser extension. | No. Not distributed as a Chrome extension. | No. No Chrome extension. | Yes, requires free plan. Chrome extension for bot-free capture. | No. No Chrome extension. | Unknown. Not confirmed this round. | No. No official browser extension. | No. No Chrome extension. | Unknown. Not confirmed this round. | No. Comet is not distributed as a Chrome Web Store extension. | No. No official browser extension. | No. No Chrome extension is documented. | No. No Chrome extension. | No. No Chrome extension. | Unknown. Not confirmed on the homepage fetch. | No | No. No Chrome extension is documented. | Unknown. Not confirmed on the homepage. | No. No official Chrome extension. | No. Not a browser extension. Do not conflate with Copilot. | No. No Chrome extension is documented. | No | Unknown. Not confirmed this round (Cloudflare). | No. Auto populated | No. No Chrome extension. | No. No Chrome extension. | No. No Chrome extension. | Unknown. Not confirmed this round. | No. No Chrome extension is documented. | No. No Chrome extension. | No. No Chrome extension. | Yes, requires free plan. Genspark ships a Chrome extension with a sidebar for multi-turn chat, page summarization, and browser automation (click, fill forms, extract data) on any webpage. Install from the Chrome Web Store; usage consumes Genspark credits. | Unknown. Overlapping columns disagreed (no vs yes); left unverified rather than copying a no. | No | Yes, requires free plan. Official Chrome extension is a primary surface. | No. No Chrome extension. | Unknown. Extension strategy was not checked; Grammarly has one separately. | No. No official browser extension. | No. No Chrome extension. | No. No Chrome extension advertised. | Yes | No. Web audio tool, not a Chrome Web Store extension. | No | No | No | No | No. Fliki is a browser-only text-to-video workspace. Official FAQ: no download, no plugin, and no GPU required — there is no Chrome extension. | No | No | No | No. No live extension from the official URL. | No | No. Auto populated | No | Yes. HighLevel documents the Ask AI LeadConnector Chrome extension. It lets Ask AI open pages, click, type, and scroll in Chrome; install from Ask AI when a browser task is requested, or from the Chrome Web Store listing. |
| 🔒 Privacy | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
🔒 GDPR Compliant | Yes. Enterprise docs link a GDPR-compliant DPA and state Cursor maintains GDPR compliance alongside SOC 2 Type II. Individual plans follow the Privacy Policy, not the DPA. | No. Auto populated | Yes | Yes | Yes | Unknown. Needs a read of Anthropic’s current DPA before we assert GDPR. | Yes | Yes. Canva operates under GDPR and publishes a DPA. | Unknown. Needs a read of the current policy before we assert this. | Unknown. Needs a current Hugging Face DPA read. | Yes. OpenAI operates under GDPR and publishes a DPA. | Yes | Yes. Auto populated | Yes | Yes. Runway publishes a DPA and serves EU customers. | Unknown. Needs a current Stability DPA read. | Unknown. Security page exists; a GDPR assertion was not copied from marketing. | Unknown. SOC 2 Type 2 is documented. A GDPR DPA was not read. | Unknown. Needs a read of the current policy before we assert this. | No | Yes. Miro operates under GDPR and publishes a DPA. | Yes | Yes. HeyGen publishes a DPA and serves EU customers. | Unknown. Needs a current Google DPA read for Gemini Notebook specifically. | Yes. Codeium publishes a DPA and serves EU customers. | Yes | Yes | Partial. api.eu.bfl.ai is documented as the EU/GDPR regional endpoint. That is data routing, not a published DPA we read here. | Yes, requires free plan. Zapier help states it complies with GDPR, UK GDPR, and CCPA, and points to the Data Privacy Overview, DTIA, and Legal pages. A DPA is published at zapier.com/legal/data-processing-addendum. | Yes. Adobe operates under GDPR across its Creative Cloud products. | Yes. Auto populated | No. Auto populated | Unknown. Local runs stay on your machine. Hosted APIs need that vendor’s DPA. | Unknown. Needs a current Replit DPA read. | Unknown. Needs a read of the current policy before we assert this. | Unknown. Needs a read of the current policy before we assert this. | Unknown. Needs a current DPA/privacy-policy read, not a guess from the model pages. | Yes | Unknown. Local-first helps, but a GDPR statement was not on the homepage. | Yes. Whimsical publishes a DPA and serves EU teams. | Unknown. SaaS vs self-host. A DPA was not read this round. | Unknown. Needs a read of the current policy before we assert this. | Unknown. Needs a current Google DPA read for Antigravity specifically. | Unknown. Needs a read of the current policy before we assert this. | Unknown. SOC 2 Type II is claimed. A GDPR statement was not isolated on the homepage. | Unknown. Needs a current Ideogram DPA read. | Yes. Auto populated | Unknown. Needs a current DPA read. Not asserted from the marketing homepage. | Yes. Eraser publishes a DPA and serves EU teams. | Unknown. Needs a current Google DPA read for Flow/Gemini video, not a guess from the model page. | Unknown. SOC 2 is documented. A GDPR DPA was not read. | Yes | Yes, requires free plan. Homepage claims GDPR. | Unknown. Needs a read of the current policy before we assert this. | Unknown. Needs an AWS DPA read for Kiro specifically. | Unknown. Needs a current DPA read. | Yes, requires free plan. Pricing page lists GDPR (and SOC 2 Type II, HIPAA). | Unknown. You send code to whichever LLM you configure. No Aider DPA was read. | Unknown. Needs a current DPA read. | Unknown. Needs a read of the current policy before we assert this. | Partial. Google Cloud compliance offerings. Specific Vertex DPA not re-read. | Unknown. Adobe privacy exists; not re-read this round. | Unknown. Needs a read of Perplexity’s current DPA before we assert GDPR for Comet. | Unknown. Needs a read of the current policy before we assert this. | Partial. Hosted plans store data in Frankfurt, Germany. Self-host wherever you choose. A DPA was not read. | Unknown. Privacy policy was not re-read. | Yes, requires enterprise plan. AWS lists GDPR, HIPAA, and FedRAMP coverage for Bedrock. | Yes, requires free plan. Homepage lists GDPR (plus SOC 2 Type II and HIPAA). | Yes | Unknown. Needs a DPA read. Data policies exist but are not GDPR certification. | Unknown. Needs a current DPA read. | Partial. Enterprise lists an EU SaaS deployment. That is data residency, not a DPA we read here. | Yes, requires plus plan. Azure OpenAI is an Azure service under the Microsoft Products and Services DPA; Trust Center documents GDPR for Azure. | Unknown. Needs a current Groq DPA read. | Yes | Unknown. Not checked. | No. Auto populated | Yes, requires plus plan. SOC2, GDPR, and ISO certifications are advertised. | Unknown. Needs a current DPA read. | Unknown. Enterprise compliance docs were not re-read. | Unknown. Needs a current DPA read. Homepage timed out. | Yes, requires plus plan. Pricing lists GDPR, SOC 2, and ISO 27001 among enterprise-grade compliance. | Unknown. Compliance docs were not re-read. | Unknown. Compliance docs were not re-read. | Partial. Genspark lists GDPR compliance as a program in progress for both Team and Enterprise. SOC 2 Type II and ISO 27001:2022 are already certified; a custom DPA is Enterprise-only. | Yes | Yes | Unknown. Needs a DPA read. | Unknown. Privacy policy was not re-read this round. | Unknown. Compliance docs were not re-read. | Unknown. Needs a read of the current policy before we assert this. | Unknown. Privacy policy was not re-read. | Unknown. Enterprise security is mentioned; a DPA was not read. | Yes | Unknown. Adobe privacy docs were not re-read. | Yes | Yes | Yes | Yes | No | Yes | Yes | No | Unknown. No live privacy statement at tome.app (404). | Yes | Yes. Auto populated | Yes | Yes |
🚫 Opt Out of Data Training | Yes. Privacy Mode is a settings toggle: with it on, Customer Data is not used for training by Cursor or model providers under ZDR. Teams/Enterprise can enforce it org-wide. | No. Auto populated | Yes | Yes | Yes | Yes, requires pro plan. Individual plans list model-training opt-out; Team and Enterprise default to no training. | Yes | Yes. Account setting controls whether your content is used to improve AI models. | Unknown. Needs a read of the current policy before we assert this. | Unknown. Public repos are public. Private-repo training policy was not checked. | Partial. Controlled through OpenAI account data controls rather than a Sora-specific toggle. | Yes | Yes. Auto populated | Yes | Unknown. Needs a read of the current policy before we assert this. | Unknown. Training use of API prompts was not checked on a privacy policy this round. | Yes, requires enterprise plan. Admins control AI content training; powered by off-the-shelf plus fine-tuned models. | Yes, requires free plan. Homepage: does not sell/monetize user content or allow third-party providers to train on it. | Unknown. Needs a read of the current policy before we assert this. | Yes | Partial. Admin controls govern AI data use on enterprise plans. | No | Unknown. Needs a read of the current policy before we assert this. | Unknown. Workspace/admin controls exist at Google generally; a Notebook-specific training opt-out was not checked here. | Yes. Zero-data-retention mode is available; enterprise plans do not train on your code. | Yes | Yes | Unknown. Training use of API prompts was not checked on a privacy policy this round. | Yes, requires free plan. Security & compliance FAQ: Enterprise is automatically opted out of AI training; other plans can opt out via Zapier's opt-out form. MCP docs repeat Enterprise default opt-out under AI Supplementary Terms. | Yes. Firefly is not trained on customer content, and Creative Cloud offers a content-analysis opt-out. | Yes. Auto populated | No. Auto populated | Unknown. Local inference sends nowhere. Hosted API training policies vary by vendor. | Unknown. A training opt-out toggle was not checked. | Unknown. Needs a read of the current policy before we assert this. | Unknown. Needs a read of the current policy before we assert this. | Unknown. Training opt-out was not found on the credits policy or VIDEO 3.0 comparison guide. | No | Partial. Context and skills stay on your computer. The models you sign into still follow those vendors’ training policies. | Unknown. Needs a read of the current policy before we assert this. | Partial. Self-host / BYOC keeps data on your infra. SaaS training policy was not checked. | Unknown. Needs a read of the current policy before we assert this. | Unknown. Training opt-out was not on the plans page checked. | Unknown. Needs a read of the current policy before we assert this. | Yes, requires free plan. Homepage: agreements ensure zero data training and retention on all third-party model providers. | Unknown. Training use of prompts was not checked on a privacy policy this round. | Yes. Auto populated | Unknown. Training-use toggle was not checked on a privacy policy this round. | Unknown. Needs a read of the current policy before we assert this. | Partial. Paid Gemini API Veo 3.1 lists “Used to improve our products: No”. Consumer Flow/Gemini training controls were not checked here. | Yes, requires free plan. Pricing: Zero Data Retention with contracted LLM providers — no customer data retained or used for training. Telemetry is configurable. | No | Unknown. Training-use policy was not on the pricing page. | Unknown. Needs a read of the current policy before we assert this. | Unknown. Enterprise privacy is advertised; a consumer training opt-out toggle was not checked. | Unknown. Docs mention data-retention controls; the toggle was not re-read this round. | Yes, requires free plan. Pricing page: they do not train on your data. | Unknown. Training policy is the LLM vendor’s, plus whatever you self-host. | Unknown. Agent training policy was not on the pricing page. | Unknown. Needs a read of the current policy before we assert this. | Partial. Enterprise terms restrict training on customer data. Consumer Gemini policies differ. | Unknown. Firefly vs Express training policy was not re-read. | Unknown. Training opt-out is not documented on the Comet privacy-and-safety article we checked. | Unknown. Needs a read of the current policy before we assert this. | Unknown. Self-host keeps data on your infra. Cloud training policy was not checked. | Yes, requires pro plan. Raycast states it does not train on your data. | Yes, requires plus plan. AWS states it does not train on customer content submitted to Bedrock. | Unknown. Training policy was not on the homepage snippet. | Yes | Partial. Custom data policies restrict which providers see prompts. Not a single training-opt-out toggle we read. | Unknown. Training policy was not on the homepage. | Unknown. A consumer training opt-out toggle was not checked. | Yes, requires plus plan. Prompts, completions, embeddings, and training data are not used to train foundation models without customer permission. | Unknown. Training use of API prompts was not checked on a privacy policy this round. | No | Unknown. Not checked. | No. Auto populated | Unknown. Training on user media was not checked. | Unknown. Training use of prompts was not checked. | Partial. Your vectors stay in your index. Training policy on hosted models was not checked. | Unknown. Not checked. | Yes, requires plus plan. Zero code retention: no storage, no training on your code, no sharing with third parties. | Unknown. Training on customer audio was not checked. | Unknown. Training policy was not checked. | Yes, requires plus plan. Team and Enterprise seats are automatically opted out of model training. Genspark says prompts and content are never used to train its models, and subprocessors are bound by no-training and Zero Data Retention agreements where applicable. | No | No | Unknown. Not on the homepage. | Unknown. Training use of uploaded faces was not checked. | Unknown. Training policy was not checked. | Unknown. Needs a read of the current policy before we assert this. | Unknown. Training on uploads was not checked. | Unknown. Not on the homepage. | Yes | Unknown. Training policy was not checked. | Yes | Yes | Yes | No | No | Yes | Yes | No | Unknown. No live privacy controls at tome.app (404). | No | Yes. Auto populated | Yes | Yes |
🏢 Enterprise Privacy Mode | Yes, requires enterprise plan. Enterprise adds Privacy Mode enforcement, ZDR, optional CMEK, data residency, SSO/SCIM, and a DPA. Privacy Mode is on by default for Enterprise teams. | No. Auto populated | Yes, requires enterprise plan | Yes, requires enterprise plan | Yes | Yes, requires enterprise plan. Team and Enterprise add SSO, SCIM, audit logs, custom retention, and a HIPAA-ready option. | Yes, requires enterprise plan | Yes, requires enterprise plan. Canva Enterprise adds SSO, admin controls and data governance. | Unknown. Not confirmed either way. | Yes, requires enterprise plan. Team/Enterprise from $20/user/month: security, access controls, dedicated support. | Unknown. Not confirmed either way. | No | Yes. Auto populated | Yes, requires enterprise plan | Yes, requires enterprise plan. Enterprise tier with SSO and custom data terms. | Yes, requires enterprise plan. Enterprise API plans list on-prem/self-host, SLAs, and verified-customer safety. Brand Studio Enterprise adds SSO. | Yes, requires enterprise plan. SSO, SCIM, localized hosting, Governance+ add-on. Enterprise Full $90/mo. | Yes, requires enterprise plan. SSO/SAML, SCIM, confidential mode, DLP, BYOK, audit logs. SOC 2 Type 2. | Unknown. Not confirmed either way. | No | Yes, requires enterprise plan. Enterprise plan with SSO, SCIM and data residency options. | No | Yes, requires enterprise plan. Enterprise tier with SSO and custom retention. | Yes, requires enterprise plan. Workspace business and Education accounts can use Gemini Notebook when an admin enables it. Ultra Access / AI Expanded Access get the cloud computer first. | Yes, requires enterprise plan. Self-hosted and VPC deployment options for enterprise. | Yes, requires enterprise plan | Yes | Yes, requires enterprise plan. Pricing lists Enterprise self-host licensing plus custom volume/SLA. EU and US regional APIs exist. | Yes, requires enterprise plan. Enterprise: SCIM, advanced admin, custom retention, observability. Team already has SAML SSO. | Yes, requires enterprise plan. Enterprise tier with IP indemnification and content credentials. | Yes. Auto populated | No. Auto populated | Partial. Self-host for isolation. No Whisper Enterprise SKU on GitHub. | Yes, requires enterprise plan. SSO/SAML, SOC 2, admin controls, single-tenant, static outbound IPs. | Unknown. Not confirmed either way. | Unknown. Not confirmed either way. | Partial. Credits policy defines Team Memberships and Enterprise Users with admin-allocated credits. That is not a published ZDR/SSO suite. | Yes, requires enterprise plan | Partial. Windows sandbox/execution-container quotes exist. There is no OpenClaw Enterprise SKU on the homepage. | Yes, requires enterprise plan. Organisation plan with SSO and admin controls. | Yes, requires enterprise plan. Govern: LLM Gateway, spend limits, sensitive-data detection. Deploy SaaS, BYOC, or self-hosted. 5 of Fortune 10. | Yes, requires enterprise plan. Enterprise plan with SSO and custom data handling. | Yes, requires enterprise plan. Teams use Google Cloud terms via Gemini Enterprise Agent Platform. Individual BYOK is explicitly unsupported. | Unknown. Not confirmed either way. | Yes, requires enterprise plan. SSO, SOC 2 Type II, 24/7 support. Seat dollars were not extracted. | Yes, requires enterprise plan. Docs send enterprise/private workflows and API scale to partnership@ideogram.ai. | Yes. Auto populated | Yes, requires enterprise plan. Enterprise adds SSO, SCIM, and a HIPAA add-on on the pricing page. | Yes, requires enterprise plan. Enterprise plan with SSO and admin controls. | Yes, requires enterprise plan. DeepMind lists Gemini Enterprise Agent Platform as a Veo surface for governed production use. | Yes, requires enterprise plan. SAML SSO, ZDR, BYOLLM, self-hosted cloud agents, admin data controls. | No | Yes, requires enterprise plan. SOC 2 Type 2 claimed. API/enterprise for catalogs. | Unknown. Not confirmed either way. | Yes, requires enterprise plan. IAM, SSO, usage dashboards, IP indemnity, GovCloud, centralized billing. | Yes, requires enterprise plan. Enterprise plans, logging/monitoring, scale. Details are sales, not a public seat price. | Yes, requires enterprise plan. SSO, SOC 2, HIPAA listed. Seat dollars were not extracted from the public table. | Partial. You can point it at local/self-hosted models so code never leaves the machine. No Aider enterprise SKU. | Yes, requires enterprise plan. Enterprise: SAML, SCIM, IP restrictions, audit log, HIPAA listed. | Unknown. Not confirmed either way. | Yes, requires enterprise plan. VPC-SC, CMEK, and Gemini Enterprise privacy controls. | Unknown. Adobe enterprise exists; Express-specific terms were not on the sign-in page. | Partial. Comet supports Chrome-like MDM policies (extensions, cookies, data clearing). That is not a published ZDR/SSO suite. | Unknown. Not confirmed either way. | Yes, requires enterprise plan. On-prem/air-gap, SSO SAML/LDAP, secret stores, RBAC, audit logs, SIEM streaming. | Partial. Team plans exist. Enterprise DPA not read. | Yes, requires enterprise plan. VPC endpoints, KMS encryption, Guardrails, and enterprise AWS agreements. | Yes, requires enterprise plan. SSO/SCIM plus SOC 2 / GDPR / HIPAA on the homepage. Used at 300k+ companies. | Yes. Self-hosted | Yes, requires plus plan. Org accounts and fine-grained data policies for trusted providers. | Yes, requires enterprise plan. Every agent has identity, scoped permissions, audit trail, and cost constraints. | Yes, requires enterprise plan. Custom RBAC, SSO, audit logging, self-hosting, vendor security review. | Yes, requires enterprise plan. Private VNet, customer-managed keys, and Microsoft enterprise agreements. | Unknown. Enterprise contracts exist in market; specifics were not on the public pricing page fetched. | No | Unknown. Not checked. | No. Auto populated | Yes, requires enterprise plan. Enterprise plans with compliance certifications. | Unknown. Dedicated endpoints exist; enterprise contract terms were not on the homepage. | Yes, requires enterprise plan. SOC 2, private networking, and enterprise plans. | Unknown. Enterprise plans exist in market; not confirmed this round. | Yes, requires plus plan. SaaS, VPC, on-prem, or fully air-gapped. SSO, encryption, IP indemnification, license-safe usage. | Yes, requires enterprise plan. Self-hosted and private cloud deployment options. | Yes, requires enterprise plan. Guardrails, EU data residency options, and enterprise contracts. | Yes, requires plus plan. Team and Enterprise include SOC 2 Type II, ISO 27001, encryption in transit and at rest, Zero Data Retention with subprocessors where applicable, and no training on customer data. Enterprise can add a custom DPA, data residency (US/EU/APAC), and a dedicated VPC. | Yes, requires enterprise plan | No | Unknown. Not on the homepage. | Partial. Enterprise plans exist. Specific DPA not read. | Partial. Enterprise mail plans exist. DPA not read this round. | Unknown. Not confirmed either way. | Unknown. Enterprise terms were not read. | Yes, requires enterprise plan. SSO, advanced permissions, brand guardrails for larger enterprises. | Yes, requires enterprise plan | Unknown. Enterprise terms were not read. | No | No | Yes, requires enterprise plan | Yes, requires enterprise plan | No | No | No | No | Unknown. No live enterprise SKU at tome.app (404). | No | Yes. Auto populated | No | Yes. HIPAA compliant (add-on) |
| 💼 Workspaces | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
📂 Projects / Folders | Yes. You open a project folder after install. Agents Window adds multi-workspace; Agent search docs cover multi-root workspaces with per-folder context. | No. Auto populated | Paid Only. Auto populated | Yes. Projects feature | Yes | Yes, requires pro plan. Each repository is a workspace; CLAUDE.md and settings travel across surfaces. | Yes. Auto populated | Yes. Designs are organised into folders and brand kits. | Unknown. Not confirmed either way. | Yes, requires free plan. Orgs, collections, and git repos organize work. | Unknown. Not confirmed either way. | Yes | No. Auto populated | Unknown. Overlapping columns disagreed (paid_only vs yes); left unverified rather than copying a no. | Yes. Assets and sessions are organised per project. | Partial. Brand Studio Enterprise lists project access controls. Not a notebook product. | Yes, requires plus plan. Unlimited files/folders on Professional. Starter is limited. | Partial. Style guide, brand tones, and snippets persist. Not a chat project workspace. | Yes. Each indexed repository is its own graph. | No | Yes. Work is organised into files, boards or folders. | Yes. Collections/Spaces | Yes. Videos are organised into projects and folders. | Yes, requires free plan. Each notebook is a project with its own sources. Auto-label kicks in at 5+ sources. | Yes. Organised by workspace. | Partial. Gems and conversation history | Yes | No. No notebook/project product. Dashboard jobs only. | Yes, requires plus plan. Shared folders and Zap workflows on Team. Free is solo. | Yes. Assets are organised in Creative Cloud libraries. | No. Auto populated | No. Auto populated | No. No project product. | Yes, requires free plan. Workspaces/projects. Core lists unlimited workspaces. | Unknown. Not confirmed either way. | Unknown. Not confirmed either way. | Unknown. Team/personal spaces exist for credits. A foldered project model was not confirmed. | Paid Only | Yes, requires free plan. Memories, skills, cron, and machines persist on your hardware. | Yes. Work is organised into files, boards or folders. | Yes, requires plus plan. LangSmith projects, datasets, and deployments. | Yes. Work is organised into projects and workspaces. | Yes, requires free plan. Antigravity 2.0 groups conversations into Projects and works across workspaces. | Unknown. Not confirmed either way. | Yes, requires free plan. Spaces, folders, lists, and projects. | Unknown. Folder/project organization in the app was not confirmed on the getting-started page. | Yes. Auto populated | Partial. Meetings can be organized; workspace folders on paid were not fully mapped. | Yes. Work is organised into files, boards or folders. | Unknown. Flow has studio projects; a Veo-specific folder model was not confirmed on the DeepMind page. | Yes, requires free plan. Warp Drive (limited on Free) plus factory config repos. | No | Partial. Brand Kit and batches. Folder UX not fully mapped. | Yes. Generations are organised into projects on the canvas. | Yes, requires free plan. Workspace .kiro/specs and steering; Crew is a persistent local workspace. | Unknown. Overlapping columns disagreed (no vs partial); left unverified rather than copying a no. | Unknown. Overlapping columns disagreed (partial vs yes); left unverified rather than copying a no. | Yes, requires free plan. The git repo is the project. Codebase map covers the tree. | Yes, requires free plan. Projects, cycles, initiatives. Teams limited on Free (2). | Unknown. Not confirmed either way. | Unknown. Overlapping columns disagreed (yes vs partial); left unverified rather than copying a no. | Partial. Saved files in an Adobe account. Not a coding project environment. | Unknown. Browser profiles exist; Perplexity Spaces inside Comet were not confirmed. | Yes. Generations are grouped into collections. | Yes, requires plus plan. Shared projects (1 on Starter, more on higher plans). | Partial. Extensions and snippets organize workflows, not chat project folders. | Partial. Knowledge Bases, agents, and guardrails are organized in the console. Not chat project folders. | Yes, requires free plan. Call library persists. | No | Unknown. Org/app keys exist. Folder workspaces were not confirmed. | Yes, requires free plan. Projects, portfolios, and goals. | Yes, requires free plan. You pick repositories; Team adds multi-repo analysis. | Yes, requires plus plan. Microsoft Foundry organizes work into projects and resources in the Azure tenant. | No. No project/folder product in the API. | No | Unknown. Not confirmed. | No. Auto populated | Yes, requires plus plan. Projects organize videos and assets. | No. No consumer project product. | Yes, requires plus plan. Projects organize indexes and Assistant apps. | Yes, requires free plan. Scenarios persist in the cloud. | Yes, requires pro plan. Unlimited codebase connections (GitHub, GitLab, Bitbucket, Perforce) on the Agentic Platform. | Partial. Console projects for API keys and usage. | Partial. Dashboard organizes API keys and usage. | Unknown. Workspace 6.0 implies projects; folder UX was not documented. | Yes. Campaigns and projects | No | No. Not a project product. | Yes, requires plus plan. Projects organize videos and agents in studio. | Partial. Labels and splits organize mail; docs have folders. | Unknown. Not confirmed either way. | Partial. Web account may store generations. | Yes, requires plus plan. Shared libraries, themes, and team workspaces. | No | Unknown. Project folders were not observable. | Yes | No | Partial. Auto populated | Yes. Workspaces | No | Yes | No | No | No. No live product at tome.app. | No. Auto populated | Yes. Auto populated | No. Auto populated | Yes. Sub-accounts and workspaces |
Showing 41 features across 108 tools.
Browse the directory
Browse all 1,374 toolsRecently added
Adobe Firefly VideoAdobe’s commercially-safe video model — not Firefly image, not PremiereAdded 2026-09-09video_generationmodel_basedai_videoenterpriseArchifyVerified architecture maps as an agent skillAdded 2026-09-09agent-skillopen-sourcediagrammingdevtoolsast-grepStructural search and rewrite over the ASTAdded 2026-09-09cliopen-sourcedevtoolscontext-layerClipchampMicrosoft browser and Windows video editor — Copilot lives inside itAdded 2026-09-09video_editingvideomicrosoftwindowsCodexOpenAI’s terminal coding agentAdded 2026-09-09coding-agentopen-sourceclidevtoolsContext7Up-to-date library docs inside the agent promptAdded 2026-09-09mcp-serveropen-sourcedocsdevtoolsDreaminaByteDance international gen studio — not CapCutAdded 2026-09-09video_generationimage_generationstudioai_videoGemini OmniMultimodal video gen and conversational edit — not Veo 3.1Added 2026-09-09video_generationvideo_editingmultimodalmodel_based
DevTools 124
View all 124AgentOpsAgentOps is an observability platform for AI agents that records session replays, tracks token costs, and surfaces errors across agent runs. Developers building multi-agent systems use it to understand why an agent failed, how long each step took, and how much a session cost. It integrates with frameworks like LangChain, CrewAI, and AutoGen through a simple SDK.Added 2026-06-15developer_toolscode_generationinfrastructuredeploymentAmazon BedrockAWS's managed service for accessing foundation models via APIAdded 2025-09-21developer_toolscode_generationinfrastructuredeploymentAMD ROCmOpen software platform for AI development and GPU computingAdded 2026-07-17developer_toolscode_generationinfrastructuredeployment
Code Assistant 43
View all 43AICodeAICode is a coding assistant built for large, complex legacy codebases rather than fresh projects. It understands multi-file dependencies and systems where context spans hundreds of files, helping developers navigate, refactor, and add features without breaking existing behavior. The focus is on enterprise teams working on mature software with years of decisions baked in.Added 2026-06-21code_generationdeveloper_toolsproductivitycodingAideAI-native code editor built on VS CodeAdded 2025-06-15code_generationdeveloper_toolsproductivitycodingAiderTerminal-based AI pair programmer for your git repoAdded 2025-01-01code_generationdeveloper_toolsproductivitycoding
Writing Assistants 43
View all 43AI-WriterVerifiable AI article generator with source citationsAdded 2025-09-21writingtext_generationproductivitycontentAIWriteBookAll-in-one AI book creation platform used by 15,000+ authorsAdded 2025-01-01writingtext_generationproductivitycontentArticle ForgeAutomated long-form article generation from keywordsAdded 2025-09-21writingtext_generationproductivitycontent
Video Generator 39
View all 39AnimotoAI-assisted video creation from photos and clipsAdded 2025-06-15video_generationcreativetext_to_videoanimationBiteableFast AI video maker for business contentAdded 2025-09-21video_generationcreativetext_to_videoanimationFlikiTurn text into videos with AI voices and visualsAdded 2025-09-21video_generationcreativetext_to_videoanimation
Explore AI Tools
Midjourney
Midjourney
Photorealistic image generation from natural text
1 of 41 features supported
View DetailsGitHub Copilot
GitHub
AI pair programmer integrated into popular IDEs
22 of 41 features supported
View DetailsChatGPT
OpenAI
Versatile conversational AI for everyday work
35 of 41 features supported
View DetailsElevenLabs
ElevenLabs
Lifelike speech synthesis and voice cloning technology
16 of 41 features supported
View DetailsClaude Code
Anthropic
Agentic coding in the terminal, IDE, desktop app, and browser
26 of 41 features supported
View DetailsNotion AI
Notion
Integrated knowledge management and document synthesis
22 of 41 features supported
View DetailsCanva Magic Studio
Canva
All in one graphic suite powered by AI
18 of 41 features supported
View DetailsSuno
Suno AI
Full song composition with vocals from text
5 of 41 features supported
View DetailsHugging Face
Hugging Face
Model hub, datasets, Spaces, and inference
22 of 41 features supported
View DetailsSora
OpenAI
Text-to-video AI model (limited access)
9 of 41 features supported
View DetailsGamma
Gamma App
AI generated interactive slide decks and documents
14 of 41 features supported
View Detailsv0
Vercel
Generative UI development platform for web components
17 of 41 features supported
View DetailsClaude
Anthropic
Advanced reasoning and nuanced long form analysis
25 of 41 features supported
View DetailsRunway
Runway AI
Cinematic video generation and visual effects
14 of 41 features supported
View DetailsStable Diffusion
Stability AI
Open image models, API, and Brand Studio from Stability AI
13 of 41 features supported
View DetailsFigma
Figma
Collaborative design canvas with Figma AI, Make, and MCP
21 of 41 features supported
View DetailsGrammarly
Grammarly
AI writing assistant for grammar, tone, and rewrites
15 of 41 features supported
View DetailsGraphify
Graphify Labs
Knowledge graph layer for AI coding assistants
4 of 41 features supported
View DetailsDeepSeek
DeepSeek AI
Open weights reasoning model for complex computation
19 of 41 features supported
View DetailsMiro AI
Miro
Collaborative whiteboarding with automated diagram generation
18 of 41 features supported
View DetailsPerplexity AI
Perplexity AI
Answer engine with real-time web search
15 of 41 features supported
View DetailsHeyGen
HeyGen
Realistic AI avatar videos from script prompts
14 of 41 features supported
View DetailsNotebookLM
Gemini Notebook — research grounded in your sources
22 of 41 features supported
View DetailsWindsurf
Codeium
Agentic IDE designed for deep codebase collaboration
20 of 41 features supported
View DetailsGoogle Gemini
Multimodal AI integrated into Google Workspace
34 of 41 features supported
View DetailsDescript
Descript
Transcript based podcast and video editing workspace
20 of 41 features supported
View DetailsFLUX
Black Forest Labs
Open-weight and API image generation from Black Forest Labs
12 of 41 features supported
View DetailsZapier
Zapier
No-code automation, agents, and 9,000+ app connections
20 of 41 features supported
View DetailsAdobe Firefly
Adobe
Commercially safe generative tools for creators
15 of 41 features supported
View DetailsBolt New
Vercel
AI web app generator (successor to v0)
15 of 41 features supported
View DetailsDeepL
DeepL
Neural machine translation for accurate multilingual content
5 of 41 features supported
View DetailsWhisper
OpenAI
Open-weight speech recognition from OpenAI
10 of 41 features supported
View DetailsReplit
Replit
Browser IDE and Agent that publishes full-stack apps
14 of 41 features supported
View DetailsUdio
Udio Music
High fidelity AI music generation and song editing
3 of 41 features supported
View DetailsNapkin AI
Napkin AI
Instant text to visual flowchart generation
4 of 41 features supported
View DetailsKling AI
Kuaishou
Text- and image-to-video with native audio
11 of 41 features supported
View DetailsMicrosoft Copilot
Microsoft
Everyday AI assistant powered by GPT and Bing
21 of 41 features supported
View DetailsOpenClaw
OpenClaw
Open-source personal agent on your machine and chat apps
18 of 41 features supported
View DetailsWhimsical
Whimsical
Visual collaborative workspace with AI diagramming
9 of 41 features supported
View DetailsLangChain
LangChain
Open agent framework plus LangSmith for test, deploy, and govern
23 of 41 features supported
View DetailsJulius AI
Julius AI
Conversational data analysis and automated spreadsheet charting
11 of 41 features supported
View DetailsGoogle Antigravity
Agentic IDE, CLI, SDK, and desktop command center
18 of 41 features supported
View DetailsPika
Pika Labs
Idea to video creation with custom animation controls
6 of 41 features supported
View DetailsClickUp
ClickUp
Projects, docs, chat, and Brain agents in one workspace
23 of 41 features supported
View DetailsIdeogram
Ideogram
Image generation with reliable in-image text
13 of 41 features supported
View DetailsLovable
Lovable
Full-stack web development with AI pair programmer
21 of 41 features supported
View DetailsOtter.ai
Otter.ai
AI meeting notes, live transcription, and Otter Chat
19 of 41 features supported
View DetailsEraser
Eraser
Diagrams and technical documentation for engineering teams
8 of 41 features supported
View DetailsVeo 3
Google DeepMind
Text- and image-to-video with native audio
11 of 41 features supported
View DetailsWarp
Warp
AI terminal, agent CLI, and cloud software factories
22 of 41 features supported
View DetailsGrok
xAI
Real-time knowledge AI with access to X/Twitter
17 of 41 features supported
View DetailsPhotoroom
Photoroom
Product photo editor, background removal, and listing API
14 of 41 features supported
View DetailsRecraft
Recraft AI
Professional visual design and scalable vector graphics
9 of 41 features supported
View DetailsKiro
Amazon
Spec-driven agentic IDE, CLI, web, and mobile
21 of 41 features supported
View DetailsReplicate
Replicate
Run and fine-tune public models via API
18 of 41 features supported
View DetailsFireflies.ai
Fireflies.ai
Meeting recorder, transcripts, AskFred, and conversation intel
20 of 41 features supported
View DetailsAider
Aider
Open-source AI pair programming in the terminal
15 of 41 features supported
View DetailsLinear
Linear
Issue tracker with agents, MCP, and Linear Agent
17 of 41 features supported
View DetailsLuma Dream Machine
Luma AI
AI video generation from images and text
7 of 41 features supported
View DetailsVertex AI
Google Cloud agent platform, Model Garden, and Gemini API
30 of 41 features supported
View DetailsAdobe Express
Adobe
Quick Firefly-powered editor for social, video, and web
7 of 41 features supported
View DetailsComet
Perplexity AI
AI-native browser with a sidecar assistant
15 of 41 features supported
View DetailsLeonardo.Ai
Leonardo.AI
Photorealistic image generation and editing
11 of 41 features supported
View Detailsn8n
n8n
Source-available workflow automation with AI agents
24 of 41 features supported
View DetailsRaycast AI
Raycast
OS launcher chat with 32+ models, extensions, and Ollama
19 of 41 features supported
View DetailsAmazon Bedrock
Amazon
AWS platform for foundation models, agents, and Guardrails
28 of 41 features supported
View DetailsFathom
Fathom
AI meeting notes, bot-free capture, and Ask Fathom
13 of 41 features supported
View DetailsLlama
Meta
Open-source large language model family
15 of 41 features supported
View DetailsOpenRouter
OpenRouter
One OpenAI-compatible API for 500+ models
18 of 41 features supported
View DetailsAsana
Asana
Work graph, AI Teammates, and MCP for human-agent teams
14 of 41 features supported
View DetailsCodeRabbit
CodeRabbit
Agentic AI code review for pull requests, IDE, and CLI
20 of 41 features supported
View DetailsAzure OpenAI Service
Microsoft
OpenAI models on Azure with enterprise identity and networking
27 of 41 features supported
View DetailsGroq
Groq
Fast LPU inference API, playground, and Whisper STT
14 of 41 features supported
View DetailsMeta AI
Meta
Open-source Llama models and Meta Advantage suite
8 of 41 features supported
View DetailsQuillBot
QuillBot
Paraphraser, grammar checker, and citation helper
2 of 41 features supported
View DetailsSynthesia
Synthesia
AI video generation with virtual presenters
5 of 41 features supported
View DetailsInVideo
InVideo
Agent Two: AI video agents, storyboard, and a timeline editor
18 of 41 features supported
View DetailsTogether AI
Together AI
Open-model inference, fine-tuning, and GPU cloud
10 of 41 features supported
View DetailsPinecone
Pinecone
Serverless vector database, embeddings, and Assistant
15 of 41 features supported
View DetailsMake
Make
Visual automation scenarios across thousands of apps
11 of 41 features supported
View DetailsTabnine
Tabnine
Private, air-gappable AI coding assistant and agents
19 of 41 features supported
View DetailsDeepgram
Deepgram
Speech-to-text, Flux TTS, and a Voice Agent API
17 of 41 features supported
View DetailsAssemblyAI
AssemblyAI
Speech-to-text, Voice Agent API, and speech understanding
16 of 41 features supported
View DetailsGenspark
Genspark
All-in-one AI workspace for search, dashboards, and agents
14 of 41 features supported
View DetailsJasper
Jasper
AI content platform for marketing and copywriting
14 of 41 features supported
View DetailsMistral AI
Mistral AI
Open-weight efficient LLM with enterprise API
12 of 41 features supported
View DetailsWordtune
Wordtune
Rewrite, tone, grammar, and summaries with a Chrome extension
10 of 41 features supported
View DetailsD-ID
D-ID
Talking avatars, agentic video, and a streaming API
21 of 41 features supported
View DetailsSuperhuman
Superhuman
Mail, docs, Grammarly, and Go — AI across every app
15 of 41 features supported
View DetailsKrea
Krea AI
Real-time AI art generation with live canvas
7 of 41 features supported
View DetailsStable Audio
Stability AI
Generate music and sound effects in the browser, DAW, or API
14 of 41 features supported
View DetailsBeautiful.ai
Beautiful.ai
AI decks with Smart Slides, themes, and team brand control
12 of 41 features supported
View DetailsYou.com
You.com
Privacy-focused search engine with AI
19 of 41 features supported
View DetailsAdobe Podcast
Adobe
Browser Enhance Speech and podcast tools from Adobe
3 of 41 features supported
View DetailsMurf AI
Murf AI
AI video generation with synthetic voices and avatars
13 of 41 features supported
View DetailsPoe by Quora
Quora
Platform for exploring and comparing AI models
18 of 41 features supported
View DetailsCohere
Cohere
Enterprise API platform for generative AI
21 of 41 features supported
View DetailsWritesonic
Writesonic
AI writing assistant for content and copy
14 of 41 features supported
View DetailsFliki
Fliki
Text-to-video and video generation with AI voices
8 of 41 features supported
View DetailsAdCreative.ai
AdCreative.ai
AI-generated ads and creative assets for marketing
15 of 41 features supported
View DetailsPi by Inflection
Inflection AI
Friendly personal AI assistant
12 of 41 features supported
View DetailsRytr
Rytr
Affordable AI writing assistant for all content types
3 of 41 features supported
View DetailsTome
Tome
Former AI presentation tool — tome.app no longer serves a product
0 of 41 features supported
View DetailsCharacter.AI
Character.AI
Chat with AI characters and personalities
15 of 41 features supported
View DetailsCopy.ai
Copy.ai
Free AI copywriting tool for marketing
18 of 41 features supported
View DetailsHuggingChat
Hugging Face
Free open-source chatbot powered by open models
17 of 41 features supported
View DetailsHighLevel
HighLevel
All-in-one business platform with AI features
24 of 41 features supported
View Details