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

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

#ModelScore
1GLM-5.3Z.ai1535
2GLM-5.3-FlashZ.ai1528
3Kimi-K3Moonshot AI1527
4GLM-5.2Z.ai1522
5Kimi-K2.6Moonshot AI1521
6GLM-5.1Z.ai1516
7Kimi-K3.1-ProMoonshot AI1516
8GLM-5Z.ai1512
9MiMo-V2.5-ProXiaomi MiMo1507
10DeepSeek-V4.1-FlashDeepSeek1505
11GLM-4.6Z.ai1500
12NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16NVIDIA1498
13Qwen3.5-397B-A17BQwen1498
14gemma-4-26B-A4B-itGoogle1495
15GLM-4.7Z.ai1494
16InklingThinking Machines1494
17Kimi-K2.5Moonshot AI1493
18Qwen3.8-27BQwen1493
19Hy3Tencent1490
20Qwen3-235B-A22B-Thinking-2507Qwen1482
Show all 141 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.