Seedance 2.0 API is now live
Gemini 3 Pro API
The Gemini 3 Pro API is Google's flagship multimodal reasoning model for long-context chat, structured outputs, and tool-aware workflows. Accept text plus image, video, or audio inputs and return high-quality text responses for production assistants and analytics.
Chat Playground
Model: gemini-3-pro
Stream
Show tokens as they arrive
Gemini 3 Pro API Complete Guide
Learn how to integrate the Gemini 3 Pro API, explore multimodal capabilities, and deploy long-context assistants in production.
What is the Gemini 3 Pro API?
The Gemini 3 Pro API is Google's flagship multimodal reasoning model. It provides developers with a REST endpoint for long-context chat, structured JSON outputs, and tool calling — no ML expertise required.
The Gemini 3 Pro API accepts text plus image, video, or audio inputs and returns high-quality text responses. Use ThinkingConfig to control reasoning depth, responseSchema for structured extraction, and tools for agent workflows.
With value pricing starting at $1.4 input and $8.4 output per 1M tokens, streaming support, and auto provider routing, the Gemini 3 Pro API delivers production-ready multimodal assistants at scale.
Why Developers Choose the Gemini 3 Pro API
Key advantages that make the Gemini 3 Pro API stand out for multimodal reasoning and long-context workflows
Gemini 3 Pro API handles long-context conversations and large outputs with stable reasoning
Multimodal input via text, image, video, or audio file parts
ThinkingConfig controls reasoning depth with includeThoughts and thinkingLevel
Structured JSON responses with responseSchema for reliable extraction
Function calling with tools and toolConfig for agent workflows
Streaming responses with auto provider routing for uptime
What Can You Build with the Gemini 3 Pro API?
From document analysis to agent workflows, the Gemini 3 Pro API powers diverse multimodal applications
Multimodal Document & Visual QA
Build pipelines that analyze documents, images, and videos. The Gemini 3 Pro API powers visual QA and document understanding with native multimodal support.
Structured Data Extraction
Extract and classify data from unstructured text with responseSchema. The Gemini 3 Pro API returns predictable JSON for reliable downstream processing.
Long-Context Knowledge Assistants
Deploy RAG-style assistants over large knowledge bases. The Gemini 3 Pro API excels at long-context reasoning and retrieval-augmented workflows.
Agent Workflows with Tool Calling
Build autonomous agents that call tools and retry. The Gemini 3 Pro API supports function calling with tools and toolConfig for production agent stacks.
Enterprise Translation & Localization
Scale translation and localization across teams. The Gemini 3 Pro API delivers high-quality multilingual outputs with clear token pricing.
Summaries & Insights from Reports
Generate executive summaries and insights from large reports. The Gemini 3 Pro API distills long documents into actionable bullet points.
Gemini 3 Pro API Pricing Snapshot
Value routing starts at $1.4 input / $8.4 output. Official routes are $2 / $12 per 1M tokens.
$1.4
per 1M tokens
$2
per 1M tokens
$8.4
per 1M tokens
$12
per 1M tokens
Value: Cost-optimized routes. Official: Direct API with highest availability.
Gemini 3 Pro API Technical Specs
Performance, input types, and output details for the Gemini 3 Pro API
Input Types
Text, Image, Video, Audio
Processing Time
1-3s
Output Type
Text, JSON
Gemini 3 Pro API Developer Reviews
Feedback from teams using the Gemini 3 Pro API in production
“Incredible multimodal reasoning. The Gemini 3 Pro API has transformed our document analysis and agent workflows entirely.”
John Smith
Senior Developer
“The Gemini 3 Pro API is dead simple to integrate and consistently delivers high-quality structured outputs. Highly recommended!”
Maria Johnson
Product Manager
“Clear API docs, fast response times, and the output quality from the Gemini 3 Pro API exceeds expectations.”
Alex Lee
Tech Lead
Gemini 3 Pro API Known Limitations
Current constraints to consider when integrating the Gemini 3 Pro API in production
responseSchema and tools are mutually exclusive in a single request
Cheap provider ignores temperature/topP/topK/maxOutputTokens parameters
thoughtSignature must be preserved in follow-up messages when present
Streaming returns SSE chunks that clients must assemble
Start Building with the Gemini 3 Pro API Today
Try the Gemini 3 Pro API in the playground above, or jump straight into the documentation to integrate it into your project.