Code with Google Gemini 3.1 Pro
in AdaL CLI
Google's elite frontier model. 77.1% ARC-AGI-2 reasoning, 80.6% SWE-bench coding, native multimodal vision, and a massive 1M token context window — all at $2 per million tokens.
No credit card required
Why developers choose Gemini 3.1 Pro
Unmatched context size, native vision, and top-tier reasoning
1M Token Context Window
Analyze entire repositories, thousands of files, and long logs in a single pass — 5x larger than Claude's 200K context.
Native Multimodal
Drop architectural diagrams, UI mockups, and error screenshots directly into the CLI. Gemini reasons across vision and code natively.
Best Price-Performance Ratio
At $2/1M input tokens, Gemini 3.1 Pro is 7.5x cheaper than Opus 4.6 while leading on most benchmarks.
#1 on ARC-AGI-2 Reasoning
Scores 77.1% on ARC-AGI-2 abstract reasoning — more than double its predecessor and 8 points ahead of Opus 4.6.
Leads 13 of 16 Benchmarks
Google reports Gemini 3.1 Pro leads on 13 of 16 evaluated benchmarks, including Terminal-Bench 2.0 and BrowseComp.
Configurable Thinking Levels
Choose Low, Medium, or High thinking per request — optimize cost for simple tasks, maximize reasoning for complex ones.
Gemini 3.1 Pro Benchmark Performance
Leading 13 of 16 evaluated benchmarks
| Benchmark | Gemini 3.1 Pro | Opus 4.6 | Sonnet 4.6 | GPT-5.2 |
|---|---|---|---|---|
| Terminal-Bench 2.0 (Agentic terminal coding) | 68.5% | 65.4% | 59.1% | 64.7% |
| SWE-bench Verified (Agentic coding) | 80.6% | 80.8% | 79.6% | 80.0% |
| ARC-AGI-2 (Abstract reasoning) | 77.1% | 68.8% | 58.3% | 52.9% |
| BrowseComp (Agentic search) | 85.9% | 84.0% | 74.7% | 65.8% |
| τ-bench Retail (Agentic tool use) | 90.8% | 91.9% | 91.7% | 82.0% |
| GPQA Diamond (Scientific reasoning) | 94.3% | 91.3% | 89.9% | 92.4% |
How to use Gemini 3.1 Pro
Get started in minutes
Install AdaL CLI
Install globally with npm. Works on macOS, Linux, and Windows.
Sign up (free)
Create your account — takes less than a minute.
Select Gemini 3.1 Pro
Use /model to pick Gemini 3.1 Pro from the model selector.
Ready to code with the best value in frontier AI?
Experience Gemini 3.1 Pro's massive context, native multimodal reasoning, and #1 benchmarks in AdaL CLI.
Get Started with AdaL