Gemini 3.1 Pro

Google DeepMind๐Ÿ‡ฌ๐Ÿ‡ง United Kingdom
active

Google's Gemini 3.1 Pro reasoning model with a thinking mode.

Context window1000K tokens
Input / 1M tokens$2
Output / 1M tokens$12

Pricing source: For prompts <= 200k tokens. Prompts > 200k tokens: input $4.00, output $18.00. Context caching $0.20/1M (<= 200k) or $0.40/1M (> 200k) plus $4.50/1M tokens/hour storage. Batch: input $1.00 (<= 200k) or $2.00 (> 200k); output $6.00 (<= 200k) or $9.00 (> 200k)..

Version History

3.1minor

Gemini 3.1 Pro enters public preview in GitHub Copilot with focus on efficient edit-then-test loops and agentic coding capabilities.

launchmajor

Gemini 3.1 Pro released. Google's Gemini 3.1 Pro reasoning model with a thinking mode.

Benchmark Scores

Full leaderboard โ†’
BenchmarkScoreRankSource
GPQA95.5%2 of 165Vals AI, Jul 2026 โ†—
SWE-bench Verified78.8%18 of 80Vals AI, Sep 2026 โ†—
MMLU-Pro91.0%1 of 76Vals AI, Sep 2026 โ†—
LiveCodeBench88.5%6 of 48Vals AI, Sep 2026 โ†—

Rank is against every model in our database with a sourced score on that benchmark, each counted once with its best score.

Coverage