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
21MiMo-V2.5Xiaomi MiMo1460
22Hy3Tencent1458
23GLM-5Z.ai1455
24gemma-4-26B-A4B-itGoogle1448
25Mistral-Large-4-1T-A52BMistral AI1446
26MiniMax-M2.7MiniMax1443
27Qwen3-VL-235B-A22B-InstructQwen1442
28Qwen3-235B-A22B-Instruct-2507Qwen1442
29DeepSeek-V4-FlashDeepSeek1442
30LongCat-Flash-ChatMeituan LongCat1441
31Inkling-SmallThinking Machines1440
32DeepSeek-V3.2DeepSeek1435
33GLM-4.5Z.ai1433
34GLM-4.6Z.ai1433
35Mistral-Medium-3.5-128BMistral AI1431
36Qwen3.5-122B-A10BQwen1431
37Qwen3-VL-235B-A22B-ThinkingQwen1429
38Qwen3.5-27BQwen1428
39MiMo-V2-FlashXiaomi MiMo1426
40Hy3-previewTencent1424
41GLM-4.7Z.ai1424
42DeepSeek-V3.2-ExpDeepSeek1423
43Mistral-Large-3-675B-Instruct-2512Mistral AI1422
44NVIDIA-Nemotron-3.5-Super-VL-120B-A12B-BF16NVIDIA1422
45Step-3.5-FlashStepFun1421
46Qwen3-Next-80B-A3B-InstructQwen1417
47granite-4.2-30bIBM Granite1408
48Qwen3.5-35B-A3BQwen1408
49DeepSeek-V3.1DeepSeek1405
50NVIDIA-Nemotron-3-Super-120B-A12B-BF16NVIDIA1399
51Qwen3-30B-A3B-Instruct-2507Qwen1396
52DeepSeek-R1-0528DeepSeek1393
53Muse-Glimmer-30Bmeta-models1388
54granite-4.2-8bIBM Granite1381
55MiniMax-M2.5MiniMax1380
56Qwen3-Next-80B-A3B-ThinkingQwen1375
57GLM-4.6VZ.ai1372
58GLM-4.5-AirZ.ai1370
59Qwen3-235B-A22BQwen1369
60Qwen3-32BQwen1362
61NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16NVIDIA1360
62Trinity-Large-ThinkingArcee AI1359
63GLM-4.7-FlashZ.ai1358
64gpt-oss-120bOpenAI1355
65NVIDIA-Nemotron-3-Nano-30B-A3B-BF16NVIDIA1355
66Ling-flash-2.0inclusionAI1353
67Trinity-Large-PreviewArcee AI1352
68GLM-4.5VZ.ai1351
69DeepSeek-V3-0324DeepSeek1350
70Ring-flash-2.0inclusionAI1343
71MiniMax-M2MiniMax1340
72Qwen3-Coder-480B-A35B-InstructQwen1338
73DeepSeek-R1DeepSeek1337
74QwQ-32BQwen1323
75INTELLECT-3Prime Intellect1323
76Qwen3-30B-A3BQwen1314
77granite-4.1-8bIBM Granite1312
78Olmo-3.1-32B-InstructAi21306
79granite-4.2-3bIBM Granite1305
80DeepSeek-V3DeepSeek1305
81gemma-3-27b-itGoogle1303
82Olmo-3.1-32B-ThinkAi21294
83Mistral-Small-3.2-24B-Instruct-2506Mistral AI1291
84Olmo-3-32B-ThinkAi21273
85DeepSeek-V2.5-1210DeepSeek1266
86gpt-oss-20bOpenAI1259
87Llama-4-Maverick-17B-128E-InstructMeta Llama1258
88Mistral-Small-3.1-24B-Instruct-2503Mistral AI1257
89granite-4.0-h-smallIBM Granite1251
90gemma-3-12b-itGoogle1247
91gemma-3n-E4B-itGoogle1246
92Qwen2.5-72B-InstructQwen1245
93Llama-3.1-405B-InstructMeta Llama1243
94Llama-3.1-Nemotron-70B-InstructNVIDIA1242
95DeepSeek-V2.5DeepSeek1239
96Llama-4-Scout-17B-16E-InstructMeta Llama1235
97Llama-3.3-70B-InstructMeta Llama1225
98gemma-3-4b-itGoogle1223
99Qwen2.5-Coder-32B-InstructQwen1220
100Llama-3.1-70B-InstructMeta Llama1208
101phi-4Microsoft1203
102Mistral-Small-24B-Instruct-2501Mistral AI1202
103aya-expanse-32bCohere Labs1182
104DeepSeek-Coder-V2-InstructDeepSeek1181
105gemma-2-27b-itGoogle1171
106Nemotron-4-340B-InstructNVIDIA1171
107Qwen2-72B-InstructQwen1171
108aya-expanse-8bCohere Labs1152
109Meta-Llama-3-70B-InstructMeta Llama1149
110gemma-2-9b-itGoogle1147
111Qwen1.5-110B-ChatQwen1144
112Llama-3.1-8B-InstructMeta Llama1144
113granite-3.1-8b-instructIBM Granite1141
114Qwen1.5-72B-ChatQwen1136
115QwQ-32B-PreviewQwen1134
116granite-3.1-2b-instructIBM Granite1131
117Qwen1.5-32B-ChatQwen1126
118Meta-Llama-3-8B-InstructMeta Llama1113
119Mixtral-8x22B-Instruct-v0.1Mistral AI1113
120Phi-3-medium-4k-instructMicrosoft1108
121gemma-2-2b-itGoogle1096
122Qwen1.5-14B-ChatQwen1094
123Llama-3.2-3B-InstructMeta Llama1090
124Mixtral-8x7B-Instruct-v0.1Mistral AI1087
125granite-3.0-8b-instructIBM Granite1087
126Phi-3-small-8k-instructMicrosoft1067
127granite-3.0-2b-instructIBM Granite1064
128Qwen1.5-7B-ChatQwen1055
129Phi-3-mini-4k-instructMicrosoft1051
130gemma-1.1-7b-itGoogle1039
131Llama-2-70b-chat-hfMeta Llama1039
132Llama-2-7b-chat-hfMeta Llama1036
133Mistral-7B-Instruct-v0.2Mistral AI1036
134Llama-2-13b-chat-hfMeta Llama1030
135Llama-3.2-1B-InstructMeta Llama1007
136gemma-7b-itGoogle1001
137Phi-3-mini-128k-instructMicrosoft984
138Qwen1.5-4B-ChatQwen980
139gemma-1.1-2b-itGoogle970
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.