gemini-3-1-proGemini 3.1 Pro is Google DeepMind’s proprietary AI model, released Feb 19, 2026. It scores 156 on the Epoch Capabilities Index. Its strongest use case is reasoning & knowledge, where it ranks in the 83rd percentile of the models Model Beat tracks.
Scores on standardized evaluations. Higher is better — and rank shows where Gemini 3.1 Pro lands among all models tracked on Model Beat.
Abstract visual reasoning
Harder abstract reasoning
Graduate-level scientific reasoning
Frontier of human expert knowledge
Common-sense trick questions
Factual accuracy & hallucination
Novel ML problem-solving
Code performance optimization
Real GitHub issue resolution
Human-rated web development
Olympiad-qualifier math
Research-level math problems
Hardest research math
Multi-step agentic tasks
Autonomous task length
Command-line agentic tasks
Gemini 3.1 Pro is a Gemini-family AI model, from Google DeepMind, released Feb 19, 2026.
Gemini 3.1 Pro is an AI model developed by Google DeepMind, part of the Gemini family, released Feb 19, 2026.
On standardized evaluations tracked by Epoch AI, Gemini 3.1 Pro scores 94.4% on GPQA Diamond, 73.5% on SimpleQA Verified, 46.4% on Humanity's Last Exam.
Gemini 3.1 Pro is a proprietary model (API access), available through its provider's API rather than as a downloadable open-weight model.
Benchmark data from Epoch AI (CC BY) and Artificial Analysis; pricing, specs & descriptions from OpenRouter. Data updated Nov 12, 2026.