The Model Index
Models/Google

Gemini 3.1 Pro

GoogleClosed weightsFlagship

Gemini 3.1 Pro is a closed AI model from Google with an index score of 161 (GPT-4 = 100), rank 46 of 217.

161146–177Index score#46 of 217 models
$2.0/ $12Price per million tokensinput / output · via OpenRouter
1.05MContext windowtokens it reads at once
19 Feb 2026Released9 months ago

Where it sitson the frontier

801001201401601802002023202420252026GPT-4 at launch = 100Gemini 3.1 Pro

GPT-5.2 reached this score 2 months earlier.

Test resultsby area · tick = best by any model

Reasoning & knowledgeevidence
GPQA Diamond (Epoch AI)
94%
Humanity's Last Exam
46%
Mathsevidence
FrontierMath (tiers 1-3)
60%
Codingevidence
SWE-bench Verified
76%
Terminal-Bench 4.0
3%
Agentsevidence
APEX-Agents
35%
Professional workevidence
Finance Agent (v2)
43%
Novel problemsevidence
ARC-AGI-2
77%
Long tasksevidence
METR time horizon
46 sec

Third-party results only, at the model’s best standard setting. Hover a bar for who ran it.

What others sayrank on 12 leaderboards

Best attop three among current models

Where to use itprices checked 18 Nov 2026

Host prices from OpenRouter; going direct can differ.

Facts and sources

Released19 Feb 2026
Announcementblog.google ↗
Release entryGemini 3.1 Pro
Scored asGemini 3.1 Pro
Output speed–
Hugging Faceclosed weights

Google releasesaround this one

This page as a chartDownload PNG