Qwen2.5
Qwen2.5 is an open-weights AI model from Alibaba with an index score of 112 (GPT-4 = 100), rank 171 of 217.
11296–128Index score#171 of 217 models
$0.36/ $0.40Price per million tokensinput / output · via OpenRouter
33KContext windowtokens it reads at once
19 Sep 2024Released2 years 2 months ago
Where it sitson the frontier
Claude 3.5 Sonnet (Jun 2024) reached this score 3 months earlier.
Test resultsby area · tick = best by any model
Reasoning & knowledgeevidence
Mathsevidence
Codingevidence
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 8 leaderboards
The Model Indexthird-party tests, our method#171of 217 · 112
LMArena · Creative writingcreative writing#140of 360 · 1353LMArena · Textoverall chat quality, judged by people#159of 362 · 1374LMArena · Long promptslong, complex requests#152of 340 · 1385LMArena · Instruction followingfollowing detailed instructions#163of 362 · 1356LMArena · Mathsmaths questions in chat#165of 346 · 1362LMArena · Codingcoding questions in chat#172of 357 · 1401Epoch Capabilities Indexoverall capability from many benchmarks#140of 231 · 131.2Best attop three among current models
Not in the top three among current models on any category we track.
Where to use itprices checked 18 Nov 2026
API providers 1 · per 1M tokens, in / out
Run it yourself
Official weightsQwen/Qwen2.5-72B-Instruct↗Host prices from OpenRouter; going direct can differ.
Facts and sources
Released19 Sep 2024
Announcementqwenlm.github.io ↗
Release entryQwen2.5
Scored asQwen2.5-72B
Output speed–
Hugging FaceQwen/Qwen2.5-72B-Instruct ↗