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

32 models ranked by LMArena score · updated Nov 11, 2026

#ModelScore
1Kimi-K3Moonshot AI1475
2GLM-5.3-FlashZ.ai1472
3DeepSeek-V4.1-FlashDeepSeek1464
4Kimi-K3.1-ProMoonshot AI1459
5Kimi-K2.6Moonshot AI1455
6Kimi-K2.5Moonshot AI1445
7InklingThinking Machines1443
8gemma-4-31B-itGoogle1443
9Qwen3.5-397B-A17BQwen1438
10Qwen3.8-27BQwen1437
11gemma-4-26B-A4B-itGoogle1434
12MiniMax-M3MiniMax1428
13MiMo-V2.5Xiaomi MiMo1427
14Qwen3-VL-235B-A22B-InstructQwen1420
15Qwen3.5-122B-A10BQwen1417
16Inkling-SmallThinking Machines1414
17Mistral-Large-4-1T-A52BMistral AI1411
18Qwen3.5-27BQwen1409
19NVIDIA-Nemotron-3.5-Super-VL-120B-A12B-BF16NVIDIA1407
20Inkling-2-SmallThinking Machines1403
21Qwen3-VL-235B-A22B-ThinkingQwen1401
22Qwen3.5-35B-A3BQwen1395
23Muse-Glimmer-30Bmeta-models1389
24GLM-4.6VZ.ai1377
25gemma-3-27b-itGoogle1358
26gemma-3-12b-itGoogle1334
27GLM-4.5VZ.ai1333
28gemma-3n-E4B-itGoogle1306
29gemma-3-4b-itGoogle1291
30Llama-4-Maverick-17B-128E-InstructMeta Llama1287
31Molmo2-8BAi21287
32Llama-4-Scout-17B-16E-InstructMeta Llama1279
Show the top 20

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.