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 coding ranked by LMArena score · updated Nov 11, 2026

#ModelScore
1GLM-5.3-FlashZ.ai1509
2DeepSeek-V4.1-FlashDeepSeek1504
3MiMo-V2.5-ProXiaomi MiMo1501
4GLM-5.3Z.ai1499
5Kimi-K2.6Moonshot AI1489
6GLM-5.2Z.ai1488
7GLM-5.1Z.ai1485
8Kimi-K2.5Moonshot AI1484
9Qwen3.8-27BQwen1479
10Beam-1.0-501B-A23Breflection1475
11MiMo-V2.5Xiaomi MiMo1469
12MiniMax-M3MiniMax1468
13NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16NVIDIA1468
14Hy3Tencent1467
15InklingThinking Machines1467
16LongCat-Flash-ChatMeituan LongCat1467
17Qwen3.5-397B-A17BQwen1463
18GLM-5Z.ai1462
19Mistral-Medium-3.5-128BMistral AI1461
20gemma-4-31B-itGoogle1459
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.