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

139 models in expert ranked by LMArena score · updated Nov 11, 2026

#ModelScore
1GLM-5.3Z.ai1519
2Kimi-K3Moonshot AI1515
3GLM-5.3-FlashZ.ai1511
4DeepSeek-V4.1-FlashDeepSeek1503
5MiMo-V2.5-ProXiaomi MiMo1502
6Kimi-K2.6Moonshot AI1492
7Kimi-K3.1-ProMoonshot AI1489
8GLM-5.2Z.ai1488
9Qwen3.8-27BQwen1479
10Inkling-2-SmallThinking Machines1478
11GLM-5.1Z.ai1477
12NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16NVIDIA1473
13InklingThinking Machines1469
14Kimi-K2.5Moonshot AI1465
15gemma-4-31B-itGoogle1465
16Beam-1.0-501B-A23Breflection1464
17DeepSeek-V4-ProDeepSeek1463
18Qwen3-235B-A22B-Thinking-2507Qwen1463
19Qwen3.5-397B-A17BQwen1462
20MiniMax-M3MiniMax1460
Show all 139 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.