Qwen3
Qwen3 is an open-weights AI model from Alibaba with an index score of 133 (GPT-4 = 100), provisional.
133110–156Index scoreProvisional · one third-party tester so far; ranked once a second confirms it
$0.12/ $0.50Price per million tokensinput / output · via OpenRouter
41KContext windowtokens it reads at once
29 Apr 2025Released1 year 7 months ago
Where it sitson the frontier
o1 reached this score 5 months earlier.
Test resultsby area · tick = best by any model
Reasoning & knowledgeevidence
Mathsevidence
Codingevidence
Agentsno evidence yet
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 7 leaderboards
LMArena · Textoverall chat quality, judged by people#90of 362 · 1434LMArena · Instruction followingfollowing detailed instructions#91of 362 · 1424LMArena · Long promptslong, complex requests#90of 340 · 1443LMArena · Mathsmaths questions in chat#92of 346 · 1433LMArena · Codingcoding questions in chat#98of 357 · 1480LMArena · Creative writingcreative writing#100of 360 · 1393Epoch Capabilities Indexoverall capability from many benchmarks#108of 231 · 142.4
Best 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/Qwen3-30B-A3B↗Host prices from OpenRouter; going direct can differ.
Facts and sources
Released29 Apr 2025
Announcementqwenlm.github.io ↗
Release entryQwen3
Scored asQwen3
Output speed–
Hugging FaceQwen/Qwen3-30B-A3B ↗