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 instructions ranked by LMArena score · updated Nov 11, 2026

#ModelScore
1GLM-5.3Z.ai1483
2GLM-5.3-FlashZ.ai1480
3MiMo-V2.5-ProXiaomi MiMo1476
4DeepSeek-V4.1-FlashDeepSeek1473
5GLM-5.2Z.ai1467
6GLM-5.1Z.ai1452
7Kimi-K2.6Moonshot AI1451
8MiMo-V2.5Xiaomi MiMo1434
9gemma-4-31B-itGoogle1433
10Qwen3.8-27BQwen1431
11Kimi-K2.5Moonshot AI1431
12GLM-5Z.ai1428
13MiniMax-M3MiniMax1428
14InklingThinking Machines1427
15Hy3Tencent1426
16Beam-1.0-501B-A23Breflection1424
17Qwen3.5-397B-A17BQwen1424
18gemma-4-26B-A4B-itGoogle1421
19NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16NVIDIA1418
20Mistral-Medium-3.5-128BMistral AI1416
21GLM-4.7Z.ai1412
22GLM-4.6Z.ai1410
23Mistral-Large-4-1T-A52BMistral AI1409
24Qwen3-235B-A22B-Instruct-2507Qwen1408
25Inkling-2-SmallThinking Machines1408
26Qwen3-VL-235B-A22B-InstructQwen1406
27MiniMax-M2.7MiniMax1405
28Mistral-Large-3-675B-Instruct-2512Mistral AI1405
29GLM-4.5Z.ai1404
30Inkling-SmallThinking Machines1402
31NVIDIA-Nemotron-3.5-Super-VL-120B-A12B-BF16NVIDIA1400
32Qwen3.5-122B-A10BQwen1399
33LongCat-Flash-ChatMeituan LongCat1397
34MiMo-V2-FlashXiaomi MiMo1393
35Qwen3.5-27BQwen1392
36Qwen3-Next-80B-A3B-InstructQwen1389
37Qwen3-235B-A22B-Thinking-2507Qwen1386
38Step-3.5-FlashStepFun1385
39Hy3-previewTencent1384
40Qwen3.5-35B-A3BQwen1380
41Qwen3-VL-235B-A22B-ThinkingQwen1374
42Muse-Glimmer-30Bmeta-models1373
43Qwen3-30B-A3B-Instruct-2507Qwen1363
44Qwen3-235B-A22BQwen1362
45Qwen3-Coder-480B-A35B-InstructQwen1355
46GLM-4.5-AirZ.ai1353
47MiniMax-M2.5MiniMax1353
48GLM-4.6VZ.ai1351
49granite-4.2-30bIBM Granite1350
50NVIDIA-Nemotron-3-Super-120B-A12B-BF16NVIDIA1347
51Qwen3-Next-80B-A3B-ThinkingQwen1344
52Trinity-Large-PreviewArcee AI1334
53MiniMax-M2MiniMax1329
54Trinity-Large-ThinkingArcee AI1326
55GLM-4.7-FlashZ.ai1326
56NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16NVIDIA1318
57Ling-flash-2.0inclusionAI1318
58gpt-oss-120bOpenAI1318
59INTELLECT-3Prime Intellect1316
60GLM-4.5VZ.ai1310
61Qwen3-32BQwen1305
62NVIDIA-Nemotron-3-Nano-30B-A3B-BF16NVIDIA1304
63Olmo-3.1-32B-InstructAi21299
64granite-4.2-8bIBM Granite1299
65QwQ-32BQwen1297
66Qwen3-30B-A3BQwen1283
67granite-4.2-3bIBM Granite1272
68granite-4.1-8bIBM Granite1269
69Llama-4-Maverick-17B-128E-InstructMeta Llama1267
70Llama-3.1-405B-InstructMeta Llama1258
71Qwen2.5-72B-InstructQwen1254
72Llama-4-Scout-17B-16E-InstructMeta Llama1248
73Llama-3.3-70B-InstructMeta Llama1242
74gpt-oss-20bOpenAI1236
75Llama-3.1-70B-InstructMeta Llama1231
76Qwen2.5-Coder-32B-InstructQwen1223
77granite-4.0-h-smallIBM Granite1222
78Llama-3.1-8B-InstructMeta Llama1159
79QwQ-32B-PreviewQwen1156
80Mixtral-8x22B-Instruct-v0.1Mistral AI1147
81granite-3.1-8b-instructIBM Granite1131
82granite-3.1-2b-instructIBM Granite1116
83Llama-3.2-3B-InstructMeta Llama1089
84granite-3.0-8b-instructIBM Granite1088
85granite-3.0-2b-instructIBM Granite1055
86Llama-3.2-1B-InstructMeta Llama1031
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.