GPT-4o
GPT-4o is a closed AI model from OpenAI with an index score of 111 (GPT-4 = 100), rank 174 of 217.
11195–127Index score#174 of 217 models
$2.5/ $10Price per million tokensinput / output · via OpenRouter
128KContext windowtokens it reads at once
13 May 2024ReleasedReplaced by GPT-4o (Aug 2024)
Where it sitson the frontier
Among the first models to reach this score.
Test resultsby area · tick = best by any model
Reasoning & knowledgeevidence
Mathsevidence
Codingno evidence yet
Agentsevidence
Professional workno evidence yet
Novel problemsno evidence yet
Long tasksevidence
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#174of 217 · 111
LMArena · Creative writingcreative writing#157of 360 · 1338LMArena · Textoverall chat quality, judged by people#184of 362 · 1346LMArena · Instruction followingfollowing detailed instructions#190of 362 · 1325LMArena · Codingcoding questions in chat#202of 357 · 1369LMArena · Mathsmaths questions in chat#198of 346 · 1309LMArena · Long promptslong, complex requests#195of 340 · 1339Epoch Capabilities Indexoverall capability from many benchmarks#144of 231 · 128.8LMArena · Visionunderstanding images#92of 145 · 1162LMArena · Searchanswering with web search#41of 41 · 1006Best attop three among current models
Replaced by GPT-4o (Aug 2024); 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
Released13 May 2024
Announcementopenai.com ↗
Release entryGPT-4o
Scored asGPT-4o (May 2024)
Output speed–
Hugging Faceclosed weights