Google19 Feb 2026Model

Gemini 3.1 Pro

Made by: Google
Released: 19 Feb 2026
Announced in: Gemini 3.1 Pro: A smarter model for your most complex tasksblog.google
Event: Gemini 3.1 Pro
API record: /api/v1/models/gemini-3-1-pro

Benchmark results

Benchmark	Score	Source check
Humanity's Last ExamNo tools · Thinking (High)	44.4%	Checked A person checked this number against its source on 16 Oct 2026. This confirms the number matches what the lab published, not that the score was reproduced. Corroborated The same number is printed in 8 other lab documents, the first 3 linked: 1 2 3
Humanity's Last ExamWith tools · search (with a blocklist) and code execution; Thinking (High)	51.4%	Checked A person checked this number against its source on 17 Oct 2026. This confirms the number matches what the lab published, not that the score was reproduced. Corroborated The same number is printed in 10 other lab documents, the first 3 linked: 1 2 3
SWE-bench Verifiedsingle attempt; Thinking (High)	80.6%	Checked A person checked this number against its source on 16 Oct 2026. This confirms the number matches what the lab published, not that the score was reproduced. Corroborated The same number is printed in 8 other lab documents, the first 3 linked: 1 2 3
SWE-bench Prosingle attempt; Thinking (High)	54.2%	Checked A person checked this number against its source on 17 Oct 2026. This confirms the number matches what the lab published, not that the score was reproduced. Corroborated The same number is printed in 16 other lab documents, the first 3 linked: 1 2 3
GPQA Diamondno tools; Thinking (High)	94.3%	Checked A person checked this number against its source on 16 Oct 2026. This confirms the number matches what the lab published, not that the score was reproduced. Corroborated The same number is printed in 9 other lab documents, the first 3 linked: 1 2 3
BrowseCompsearch, Python and browsing; Thinking (High)	85.9%	Checked A person checked this number against its source on 17 Oct 2026. This confirms the number matches what the lab published, not that the score was reproduced. Corroborated The same number is printed in 11 other lab documents, the first 3 linked: 1 2 3
ARC-AGI-2ARC Prize Verified; Thinking (High)	77.1%	Checked A person checked this number against its source on 16 Oct 2026. This confirms the number matches what the lab published, not that the score was reproduced. Corroborated The same number is printed in 3 other lab documents: 1 2 3
Terminal-Bench 2.0Terminus 2 harness; Thinking (High)	68.5%	Checked A person checked this number against its source on 16 Oct 2026. This confirms the number matches what the lab published, not that the score was reproduced. Corroborated The same number is printed in 6 other lab documents, the first 3 linked: 1 2 3
DeepSWE v1.1reported later, in Introducing Gemini 3.6 Flash, 3.5 Flash-Lite, and 3.5 Flash Cyber	12%	Checked A person checked this number against its source on 17 Oct 2026. This confirms the number matches what the lab published, not that the score was reproduced. Corroborated The same number is printed in 1 other lab document: 1

Sources

Reported by Google

Gemini 3.1 Pro: A smarter model for your most complex tasksblog.google · 19 Feb 2026
Original image
Introducing Gemini 3.6 Flash, 3.5 Flash-Lite, and 3.5 Flash Cyberblog.google · 21 Jul 2026
Original image

Events

Stories

All models