Overview
Google announced Gemini 3.1 Pro on February 19, 2026 as a Gemini 3-series model for complex problem-solving and advanced reasoning. The API model is a preview, optimized for software engineering and agentic workflows involving tools and multiple steps. (blog.google)
Capabilities and uses
- Accepts text, images, video, audio and PDFs; produces text. The documented limits are 1,048,576 input tokens and 65,536 output tokens. (aistudio.google.com)
- Supports thinking, function calling, code execution, structured output, context caching, URL context and grounding with Google Search or Google Maps. (aistudio.google.com)
- Suggested uses include explaining complex topics, synthesizing data, building dashboards, generating animated SVG code and prototyping interactive designs. Google reported a verified 77.1% on ARC-AGI-2; this benchmark is not a guarantee of performance on every task. (blog.google)
Availability and limitations
At announcement, Google described a developer preview through the Gemini API and developer tools, enterprise access through Vertex AI and Gemini Enterprise, and consumer rollout through the Gemini app and NotebookLM. Access and limits vary by product and plan. The API model does not support image generation, audio generation or the Live API. As a preview, it is subject to change; Google’s deprecation page lists no announced shutdown date for this endpoint. (blog.google)
Gemini API pricing
Google’s Gemini Developer API pricing, checked November 16, 2026. Prices below are in USD per 1 million tokens, for the paid tier of gemini-3.1-pro-preview; output includes thinking tokens. The prompt-length threshold determines the applicable tier. These are Gemini API rates, not necessarily Vertex AI or consumer subscription rates. (aistudio.google.com)
| Service | Prompt length | Input | Output |
|---|---|---|---|
| Standard | 200,000 tokens or fewer | $2.00 | $12.00 |
| Standard | More than 200,000 tokens | $4.00 | $18.00 |
| Batch or Flex | 200,000 tokens or fewer | $1.00 | $6.00 |
| Batch or Flex | More than 200,000 tokens | $2.00 | $9.00 |
| Priority | 200,000 tokens or fewer | $3.60 | $21.60 |
| Priority | More than 200,000 tokens | $7.20 | $32.40 |
For Standard, Batch and Flex, cached-context pricing is $0.20 per million tokens for prompts of at most 200,000 tokens and $0.40 above that threshold, plus $4.50 per million tokens per hour for storage. Priority caching is $0.36 and $0.72, respectively, plus $8.10 per million tokens per hour for storage. The pricing page lists no free API tier for this model; tools may incur separate charges. (aistudio.google.com)
Official sources
- Google announcement
- Gemini API model documentation
- Gemini API pricing — checked November 16, 2026.
- Gemini API deprecations