Gemini 2.5 Pro
Gemini 2.5 Pro is a closed AI model from Google with an index score of 141 (GPT-4 = 100), rank 91 of 217.
141125–157Index score#91 of 217 models
–Price per million tokensno public API price found
–Context windowtokens it reads at once
25 Mar 2025ReleasedReplaced by Gemini 2.5 Pro (May 2025)
Where it sitson the frontier
Among the first models to reach this score.
Test resultsby area · tick = best by any model
Reasoning & knowledgeevidence
Mathsevidence
Codingevidence
Agentsevidence
Professional workno evidence yet
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 12 leaderboards
The Model Indexthird-party tests, our method#91of 217 · 141
LMArena · Creative writingcreative writing#45of 360 · 1443LMArena · Instruction followingfollowing detailed instructions#71of 362 · 1438LMArena · Long promptslong, complex requests#70of 340 · 1457LMArena · Textoverall chat quality, judged by people#75of 362 · 1445LMArena · Mathsmaths questions in chat#84of 346 · 1439LMArena · Codingcoding questions in chat#112of 357 · 1464LMArena · Visionunderstanding images#49of 145 · 1247Epoch Capabilities Indexoverall capability from many benchmarks#95of 231 · 144.9LMArena · Documentsworking with long documents#33of 40 · 1423LMArena · Searchanswering with web search#34of 41 · 1143LMArena · WebDevbuilding web apps and front ends#126of 133 · 1226Best attop three among current models
Replaced by Gemini 2.5 Pro (May 2025); it no longer leads any category among current models.
Where to use itprices checked 18 Nov 2026
Host prices from OpenRouter; going direct can differ.
Facts and sources
Released25 Mar 2025
Announcementblog.google ↗
Release entryGemini 2.5 Pro
Scored asGemini 2.5 Pro (Mar 2025)
Output speed–
Hugging Faceclosed weights