Skip to main content
🔺

Vertex AI

by Google

Vertex AI is Google Cloud's fully managed machine learning platform, providing access to Gemini, PaLM, Imagen, and other Google foundation models alongside tools for fine-tuning, evaluation, and deployment. Its Model Garden offers 130+ models from Google and third parties. Vertex AI Studio provides a no-code playground for prompt engineering, while the SDK enables programmatic integration into applications. Enterprises choose Vertex AI for its deep integration with BigQuery, Google's data infrastructure, and for compliance-grade AI deployment with audit logging and VPC controls.

Vertex AI features and capabilities, including which plan they need and how they compare in the CanIUse matrix.

Used in combinations

See who owns what in this stack

More from Google

Google in the directory

Directory alternatives

Amazon BedrockAWS's managed service for accessing foundation models via APIAdded 2025-09-21developer_toolscode_generationinfrastructuredeployment#9FreemiumMatrixAzure OpenAI ServiceEnterprise-grade access to OpenAI models through Microsoft AzureAdded 2025-09-21developer_toolscode_generationinfrastructuredeployment#9FreemiumMatrixGoogle AI StudioGoogle's free developer playground for Gemini modelsAdded 2025-09-21developer_toolscode_generationinfrastructuredeployment#9FreeDeepInfraDeepInfra is a cloud inference platform for running open-source AI models through a simple API without managing GPU infrastructure. Developers use it to call models like Llama, Mistral, and Whisper at low cost, with pricing based on tokens consumed rather than reserved compute. It is a practical option for teams that want access to frontier open-source models without building their own serving stack.Added 2026-06-15developer_toolscode_generationinfrastructuredeployment#5FreemiumFireworks AIFast and affordable inference platform for open-source AI modelsAdded 2025-09-21developer_toolscode_generationinfrastructuredeployment#7FreemiumOpenRouterUnified API for 300+ AI models from every major providerAdded 2025-09-21developer_toolscode_generationinfrastructuredeployment#7FreemiumMatrixTogether AIFast cloud inference for open-source AI modelsAdded 2025-09-21developer_toolscode_generationinfrastructuredeployment#7FreemiumMatrixReplicateRun AI models in the cloud via APIAdded 2025-01-01developer_toolscode_generationinfrastructuredeployment#7PaidMatrix

✨ Modalities

💬
Text Input
Yes, requires plus plan. Gemini and other models via Generate Content API and Agent Studio.
🖼️
Image Upload / Vision
Yes, requires plus plan. Gemini multimodal accepts images in the API and console.
🎨
Image Generation
Yes, requires plus plan. Imagen models in Model Garden and Vertex AI Studio.
🎙️
Voice Mode
Partial. Speech APIs exist. Not a consumer voice-chat loop like Gemini app.
🎥
Video Understanding
Yes, requires plus plan. Gemini video understanding in Agent Studio and the API.
🎬
Video Generation
Yes, requires plus plan. Vertex AI serves Veo and other video generation models.
📁
File Upload (PDF, DOCX)
Yes, requires plus plan. RAG with Cloud Storage and document ingestion in Agent Builder.
💻
Code Execution
Partial. Colab Enterprise and notebooks for code. Not a chat Python sandbox.
🔊
Audio Input
Partial. Speech-to-text models available. Vertex is a platform, not a voice assistant.
🖌️
Canvas / Artifact Editor
No. Studio is a dev console, not a consumer canvas.
🖥️
Screen Sharing / Computer Use
No. No computer-use desktop product.

🧠 Memory & Context

🧠
Persistent Memory
Partial. RAG corpora and agent session config. Distinct from consumer Gemini memory.
📏
Large Context Window
Unknown. Depends on the Gemini/model version deployed.
⚙️
Custom Instructions
Yes, requires plus plan. System instructions in Agent Studio and API prompts.
📜
Conversation History
Partial. Session state in agents. API calls are otherwise stateless.

🔌 Integrations

🌐
Web Search / Browsing
Partial. Grounding with Google Search in Gemini API. Enterprise Agent Platform feature set.
🔌
Plugins / Extensions
Yes, requires plus plan. Extensions, tools, and Model Garden integrations.
🔧
API Access
Yes, requires plus plan. Vertex AI REST/gRPC SDK. $300 credits for new accounts.
⚡
Zapier Integration
Unknown. No first-party Zapier listing was checked.
🤖
Custom GPTs / Bots
Yes, requires plus plan. Agent Builder and Agent Studio deploy custom agents.
👥
Real-time Collaboration
Partial. IAM-shared console and notebooks. Not multiplayer chat.
🐦
Live Social Data Integration
No. No live social firehose product.
📚
Live Citation Mapping
Partial. Grounding with Search returns citations. RAG returns source chunks.

👨‍💻 Developer

🛠️
Function Calling
Yes, requires plus plan. Function calling in Gemini and agent tool use.
⚡
Streaming Responses
Yes, requires plus plan. Streaming generateContent responses.
🎛️
Fine-Tuning
Yes, requires plus plan. Supervised fine-tuning and tuning jobs in Vertex AI.
📐
Embeddings API
Yes, requires plus plan. Text embeddings and Vector Search.
📋
Structured Output (JSON)
Yes, requires plus plan. JSON schema / response schema in Gemini API.
🧩
Reasoning / Thinking Mode
Partial. Thinking/reasoning models in Gemini 2.x family. Platform-level, not a single toggle.
🏠
Local Execution
No. Cloud GCP inference. Distinct from Antigravity local dev.
🔓
Open Weights
Partial. Model Garden hosts open models (Gemma, Llama) plus Google proprietary models.
🤖
Multi-Agent Orchestration
Yes, requires plus plan. Gemini Enterprise Agent Platform supports multi-agent flows.

📱 Platforms

🌐
Web App
Yes, requires plus plan. Google Cloud console and Vertex AI Studio.
📱
iOS App
No. No consumer iOS app. Distinct from Gemini mobile.
🤖
Android App
No. No consumer Android app.
🖥️
Desktop App
No. Browser console, not a desktop chat app.
🔵
Chrome Extension
No. No Chrome extension.

🔒 Privacy

🔒
GDPR Compliant
Partial. Google Cloud compliance offerings. Specific Vertex DPA not re-read.
🚫
Opt Out of Data Training
Partial. Enterprise terms restrict training on customer data. Consumer Gemini policies differ.
🏢
Enterprise Privacy Mode
Yes, requires enterprise plan. VPC-SC, CMEK, and Gemini Enterprise privacy controls.

💼 Workspaces

📂
Projects / Folders
Unknown. Overlapping columns disagreed (yes vs partial); left unverified rather than copying a no.