The Model Index
Models/Google

Gemini 3.8 Flash

GoogleClosed weightsFlagship

Gemini 3.8 Flash is a closed AI model from Google with an index score of 173 (GPT-4 = 100), rank 21 of 217.

173157–189Index score#21 of 217 models
$0.82/ $4.13Price per million tokensinput / output · via OpenRouter
1.05MContext windowtokens it reads at once
2 Sep 2026Released3 months ago

Where it sitson the frontier

801001201401601802002023202420252026GPT-4 at launch = 100Gemini 3.8 Flash

Claude Fable 5 reached this score 3 months earlier.

Test resultsby area · tick = best by any model

Reasoning & knowledgeevidence
GPQA Diamond (Epoch AI)
95%
Humanity's Last Exam
45%
Mathsevidence
FrontierMath (tiers 1-3)
68%
Codingevidence
SWE-bench Verified
no result
Terminal-Bench 4.0
20%
Agentsevidence
APEX-Agents
64%
Professional workevidence
Finance Agent (v2)
61%
Novel problemsevidence
ARC-AGI-2
89%
Long tasksno evidence yet
METR time horizon
no result

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

Released2 Sep 2026
Announcementblog.google ↗
Release entryGemini 3.8 Flash
Scored asGemini 3.8 Flash
Output speed–
Hugging Faceclosed weights

Google releasesaround this one

This page as a chartDownload PNG