Browse 41 AI Features

Send and receive standard text messages. This is the foundational capability of all AI chat models.

Upload documents like PDFs, Word files, CSVs, or text files so the AI can read, summarize, and answer questions based on the content.

Upload images, screenshots, or photos for the AI to analyze, describe, and understand (Vision capabilities).

Provides a dedicated side-panel UI for documents and code, allowing you and the AI to collaboratively edit, highlight, and iterate on content.

Generate original artwork, photos, and diagrams directly from text descriptions using built-in image models like DALL-E or Midjourney.

Upload pre-recorded audio files or voice notes for the AI to transcribe, summarize, or analyze the speech.

The AI can write and securely execute real code (like Python) in a sandbox to perform math, analyze data, or generate charts.

Speak to the AI and hear responses back in real-time, allowing for natural, fluid vocal conversations rather than just text-to-speech.

Upload video files or link to YouTube videos for the AI to watch, summarize, and extract key information or transcripts.

Generate original video clips from text, images, or storyboards using a built-in video model — distinct from watching or editing existing footage.

The AI can view your computer screen in real-time and even take control to perform actions on your behalf (Computer Use).

Automatically saves your past chat sessions so you can resume them later, search through them, and organize your work.

Set persistent, global rules for how the AI should behave or respond across all chats (e.g., always output Python code, or never use emojis).

The AI automatically remembers important facts and preferences about you across different conversation threads over time.

The maximum amount of text (tokens) the AI can hold in its short-term memory at once. A large context (100k+) means it can read entire books in one prompt.

Provides a REST API for developers to programmatically integrate the AI model into their own custom applications and workflows.

Access to a marketplace of third-party plugins or extensions that connect the AI to external services (like Expedia, Wolfram, or OpenTable).

Multiplayer features allowing multiple human users to join the same chat session or document and interact with the AI together.

Create specialized AI agents (like Custom GPTs) with specific instructions and knowledge files, and share them with others.

The AI can browse the live internet in real-time to find up-to-date information, news, or verify facts instead of relying on outdated training data.

Native integration with Zapier to trigger automated workflows across thousands of other apps (like Gmail, Slack, or Notion).

Conversational search with live academic and web citation mapping directly in the answers.

Real-time sentiment mapping and breaking news analysis using live social data streams.

The API can stream responses token-by-token in real-time, allowing UIs to show text as it is being generated instead of waiting for the full response.

Developers can define custom tools/functions that the AI can choose to invoke, returning structured JSON arguments to execute actions.

Guarantees that the AI response will strictly adhere to a specific JSON schema, essential for reliable software engineering.

The AI enters a specialized thinking state where it spends additional time computing and planning before outputting an answer for complex math or logic.

Can orchestrate multiple sub-agents to divide complex technical problems and perform cross-reviews.

Allows developers to train the base model on their own custom datasets to specialize it for specific tasks, tones, or industries.

Can be downloaded and run locally on your own hardware without relying on cloud APIs or paying token fees.

The underlying model weights are publicly available for developers to inspect, modify, and self-host.

Access to vector embedding models, allowing developers to convert text into numerical vectors for semantic search and RAG applications.

Accessible through a standard web browser interface (like Chrome, Safari, or Firefox) without needing to install anything.

Offers a dedicated, native application for iPhone and iPad users, often including exclusive mobile features like native voice.

Offers a dedicated, native application for Android smartphones and tablets.

Offers a dedicated, native desktop application for macOS or Windows, allowing for system-level shortcuts and integrations.

Use the tool seamlessly via an official Chrome/Edge extension without leaving your current webpage.

Enterprise-grade security with zero data retention (ZDR), SSO integration, and strict data isolation for corporate clients.

Strictly complies with the European Union GDPR regulations for data protection and user privacy rights.

Provides a clear toggle or setting to ensure your conversations and uploaded files are NOT used to train future AI models.

Allows organizing separate chats, knowledge bases, and files into distinct workspaces, projects, or folders.