Gemini 3.1 Pro Preview
Gemini 3.1 Pro Preview
📌 Model Overview
Reasoning-first Gemini preview for agentic coding and complex problem solving
Gemini 3.1 Pro Preview is a Api Only model developed by Google DeepMind, released on 2026-02-19. It is designed primarily for Chat Reasoning workloads. Featuring a 1.0M tokens context window and Undisclosed parameter count, it offers robust performance for enterprise integration, developers, and researchers.
✨ Key Features & Capabilities
| Feature | Description |
|---|---|
| Context Window | 1.0M tokens capacity for extended prompts and multi-turn workflows |
| Primary Task | Optimized for Chat Reasoning |
| Deployment | api-only |
| Modality | text, image, video, audio, pdf |
| Max Output | Max Output: 66K tokens |
| Native reasoning capability | Native reasoning capability |
| Tool / function calling support | Tool / function calling support |
⚙️ Technical Specifications
| Specification | Details |
|---|---|
| Developer / Lab | Google DeepMind |
| Release Date | 2026-02-19 |
| Model Type | Api Only |
| Parameters | Undisclosed |
| Context Window | 1.0M tokens |
| License | Proprietary |
| Model Family | gemini-pro |
📊 Benchmarks & Performance
| Benchmark | Score | Source |
|---|---|---|
| SWE-Bench Pro | 54.2 (resolve rate) | Independent Eval |
| Terminal-Bench | 70.3% | Independent Eval |
| SWE-Bench Pro | 46.1 (resolve rate) | Independent Eval |
| SWE-Atlas Codebase QnA | 13.5 (score) | Independent Eval |
| SWE-Atlas Refactoring | 33.81 (score) | Independent Eval |
| SWE-Atlas Test Writing | 29.84 (score) | Independent Eval |
| Artificial Analysis Coding Agent Index | 43 (average pass@1) | Independent Eval |
| SWE-Atlas Codebase QnA | 45.6 (pass@1) | Independent Eval |
| SWE-Bench Pro | 15.1 (pass@1) | Independent Eval |
| Terminal-Bench | 68.3 (pass@1) | Independent Eval |
| GPQA Diamond | 94.3 (accuracy) | Independent Eval |
| Humanity's Last Exam | 44.4 (accuracy) | Independent Eval |
| ARC-AGI-2 | 77.1 (accuracy) | Independent Eval |
| MMMU Pro | 80.5 (accuracy) | Independent Eval |
| MCP Atlas | 78.2% | Independent Eval |
| OSWorld-Verified | 76.2% | Independent Eval |
| CharXiv Reasoning | 83.3 (accuracy) | Independent Eval |
| GDPval-AA | 1314 (Elo) | Independent Eval |
💰 Pricing
| Tier / Unit | Rate (USD) |
|---|---|
| 1M input tokens | $2 |
| 1M output tokens | $12 |
| 1M cache read tokens | $0.2 |
📜 License & Usage
This model is governed by the Proprietary license. Please check official developer guidelines before commercial deployment.
Key Features
Max Output: 66K tokens
Native reasoning capability
Tool / function calling support
You might also want to compare
Verified Sources
Tags
Model Specs
Parameters
Undisclosed
Context Window
1.0M tokens
License
Proprietary
Deployment
Pricing
- 1M input tokens$2
- 1M output tokens$12
- 1M cache read tokens$0.2
as of 2026-07-26
Lineage
Curator Notes
Skeleton entry imported from models.dev backfill (2026-07-26). Needs manual enrichment and primary-source verification.
Compare Specs
Compare parameters, context windows, modalities, and benchmark scores of this model side-by-side with others.
Compare Model