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

86 models in non-english ranked by LMArena score · updated Nov 11, 2026

#ModelScore
1GLM-5.3Z.ai1462
2GLM-5.3-FlashZ.ai1462
3GLM-5.2Z.ai1458
4DeepSeek-V4.1-FlashDeepSeek1450
5GLM-5.1Z.ai1448
6MiMo-V2.5-ProXiaomi MiMo1448
7Kimi-K2.6Moonshot AI1446
8InklingThinking Machines1436
9Kimi-K2.5Moonshot AI1432
10GLM-5Z.ai1431
11gemma-4-31B-itGoogle1430
12Qwen3.5-397B-A17BQwen1430
13NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16NVIDIA1428
14Hy3Tencent1426
15GLM-4.6Z.ai1426
16Qwen3.8-27BQwen1424
17gemma-4-26B-A4B-itGoogle1421
18GLM-4.7Z.ai1417
19GLM-4.5Z.ai1417
20Mistral-Large-3-675B-Instruct-2512Mistral AI1414
Show all 86 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.