Through an API

Open and closed models answer the same suite-standard 1.0 through their APIs, ranked together. 1 ranked model.

#ModelPass rate
1DeepSeek V4.1 Flash 1017Open weights100%

Model leaderboard

65 models in chinese ranked by LMArena score · updated Nov 11, 2026

#ModelScore
65Olmo-3.1-32B-ThinkAi21241
64granite-4.2-3bIBM Granite1279
63Olmo-3-32B-ThinkAi21299
62gpt-oss-20bOpenAI1315
61GLM-4.5VZ.ai1337
60Qwen3-30B-A3BQwen1347
59INTELLECT-3Prime Intellect1355
58Qwen3-32BQwen1357
57granite-4.2-8bIBM Granite1362
56NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16NVIDIA1365
55MiniMax-M2MiniMax1366
54Trinity-Large-ThinkingArcee AI1373
53QwQ-32BQwen1378
52NVIDIA-Nemotron-3-Nano-30B-A3B-BF16NVIDIA1385
51gpt-oss-120bOpenAI1385
50MiniMax-M2.5MiniMax1393
49DeepSeek-R1DeepSeek1400
48GLM-4.7-FlashZ.ai1403
47Qwen3-Next-80B-A3B-ThinkingQwen1406
46Muse-Glimmer-30Bmeta-models1410
Show all 65 models

Score by size

Plot against
Best for its size

Scores: LMArena, CC BY 4.0. Measured on served versions, not local builds. A model with several served variants shows its best.