Gemini 3.1 Flash-Lite
Gemini 3.1 Flash-Lite is a closed AI model from Google with an index score of 139 (GPT-4 = 100), rank 103 of 217.
139122–155Index score#103 of 217 models
$0.25/ $1.5Price per million tokensinput / output · via OpenRouter
1.05MContext windowtokens it reads at once
3 Mar 2026ReleasedReplaced by Gemini 3.5 Flash-Lite
Where it sitson the frontier
Gemini 2.5 Pro reached this score 11 months earlier.
Test resultsby area · tick = best by any model
Reasoning & knowledgeevidence
Mathsevidence
Codingevidence
Agentsevidence
Professional workevidence
Novel problemsno evidence yet
Long tasksno evidence yet
Third-party results only, at the model’s best standard setting. Hover a bar for who ran it.
What others sayrank on 10 leaderboards
The Model Indexthird-party tests, our method#103of 217 · 139
LMArena · Creative writingcreative writing#76of 360 · 1415LMArena · Mathsmaths questions in chat#87of 346 · 1436LMArena · Textoverall chat quality, judged by people#93of 362 · 1433LMArena · Instruction followingfollowing detailed instructions#112of 362 · 1409LMArena · Long promptslong, complex requests#108of 340 · 1430LMArena · Codingcoding questions in chat#124of 357 · 1458LMArena · Visionunderstanding images#58of 145 · 1235Epoch Capabilities Indexoverall capability from many benchmarks#94of 231 · 145.0LMArena · WebDevbuilding web apps and front ends#119of 133 · 1256Best attop three among current models
Replaced by Gemini 3.5 Flash-Lite; it no longer leads any category among current models.
Where to use itprices checked 18 Nov 2026
API providers 2 · per 1M tokens, in / out
Host prices from OpenRouter; going direct can differ.
Facts and sources
Released3 Mar 2026
Announcementblog.google ↗
Release entryGemini 3.1 Flash-Lite
Scored asGemini 3.1 Flash-Lite
Output speed–
Hugging Faceclosed weights