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 hard prompts ranked by LMArena score · updated Nov 11, 2026

#ModelScore
1GLM-5.3Z.ai1493
2GLM-5.3-FlashZ.ai1490
3MiMo-V2.5-ProXiaomi MiMo1485
4DeepSeek-V4.1-FlashDeepSeek1483
5GLM-5.2Z.ai1480
6GLM-5.1Z.ai1473
7Kimi-K2.6Moonshot AI1470
8Qwen3.8-27BQwen1456
9Kimi-K2.5Moonshot AI1453
10InklingThinking Machines1453
11NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16NVIDIA1452
12GLM-5Z.ai1452
13MiMo-V2.5Xiaomi MiMo1450
14Qwen3.5-397B-A17BQwen1448
15Hy3Tencent1447
16gemma-4-31B-itGoogle1447
17GLM-4.7Z.ai1443
18Beam-1.0-501B-A23Breflection1442
19MiniMax-M3MiniMax1442
20GLM-4.6Z.ai1440
21gemma-4-26B-A4B-itGoogle1439
22Mistral-Medium-3.5-128BMistral AI1436
23Mistral-Large-4-1T-A52BMistral AI1434
24Qwen3-235B-A22B-Instruct-2507Qwen1433
25LongCat-Flash-ChatMeituan LongCat1433
26Mistral-Large-3-675B-Instruct-2512Mistral AI1431
27GLM-4.5Z.ai1429
28Qwen3-VL-235B-A22B-InstructQwen1428
29Qwen3-Next-80B-A3B-InstructQwen1428
30Inkling-SmallThinking Machines1425
31MiniMax-M2.7MiniMax1422
32Qwen3.5-122B-A10BQwen1421
33Inkling-2-SmallThinking Machines1421
34MiMo-V2-FlashXiaomi MiMo1420
35Qwen3-235B-A22B-Thinking-2507Qwen1416
36Qwen3.5-27BQwen1414
37NVIDIA-Nemotron-3.5-Super-VL-120B-A12B-BF16NVIDIA1414
38Hy3-previewTencent1413
39Step-3.5-FlashStepFun1410
40Qwen3-VL-235B-A22B-ThinkingQwen1407
41Qwen3.5-35B-A3BQwen1400
42Qwen3-30B-A3B-Instruct-2507Qwen1398
43Muse-Glimmer-30Bmeta-models1395
44Qwen3-235B-A22BQwen1393
45NVIDIA-Nemotron-3-Super-120B-A12B-BF16NVIDIA1386
46GLM-4.5-AirZ.ai1379
47granite-4.2-30bIBM Granite1377
48MiniMax-M2.5MiniMax1372
49Qwen3-Coder-480B-A35B-InstructQwen1372
50Qwen3-Next-80B-A3B-ThinkingQwen1371
51GLM-4.6VZ.ai1368
52gpt-oss-120bOpenAI1363
53Ling-flash-2.0inclusionAI1362
54MiniMax-M2MiniMax1357
55GLM-4.7-FlashZ.ai1356
56INTELLECT-3Prime Intellect1352
57Trinity-Large-PreviewArcee AI1350
58NVIDIA-Nemotron-3-Nano-30B-A3B-BF16NVIDIA1348
59Trinity-Large-ThinkingArcee AI1342
60NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16NVIDIA1340
61Qwen3-32BQwen1334
62GLM-4.5VZ.ai1334
63granite-4.2-8bIBM Granite1326
64QwQ-32BQwen1325
65Olmo-3.1-32B-InstructAi21322
66Qwen3-30B-A3BQwen1314
67granite-4.2-3bIBM Granite1304
68granite-4.1-8bIBM Granite1294
69Llama-4-Maverick-17B-128E-InstructMeta Llama1281
70gpt-oss-20bOpenAI1274
71Qwen2.5-72B-InstructQwen1271
72Llama-3.1-405B-InstructMeta Llama1269
73Llama-4-Scout-17B-16E-InstructMeta Llama1266
74Llama-3.3-70B-InstructMeta Llama1257
75Qwen2.5-Coder-32B-InstructQwen1251
76Llama-3.1-70B-InstructMeta Llama1241
77granite-4.0-h-smallIBM Granite1241
78Llama-3.1-8B-InstructMeta Llama1175
79QwQ-32B-PreviewQwen1162
80Mixtral-8x22B-Instruct-v0.1Mistral AI1150
81granite-3.1-8b-instructIBM Granite1145
82granite-3.1-2b-instructIBM Granite1138
83Llama-3.2-3B-InstructMeta Llama1095
84granite-3.0-8b-instructIBM Granite1092
85granite-3.0-2b-instructIBM Granite1073
86Llama-3.2-1B-InstructMeta Llama1044
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.