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

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

#ModelScore
1GLM-5.3Z.ai1475
2GLM-5.3-FlashZ.ai1472
3GLM-5.2Z.ai1469
4MiMo-V2.5-ProXiaomi MiMo1464
5DeepSeek-V4.1-FlashDeepSeek1464
6GLM-5.1Z.ai1462
7Kimi-K2.6Moonshot AI1455
8GLM-5Z.ai1446
9NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16NVIDIA1445
10Kimi-K2.5Moonshot AI1445
11InklingThinking Machines1443
12gemma-4-31B-itGoogle1443
13GLM-4.6Z.ai1440
14Hy3Tencent1439
15Qwen3.5-397B-A17BQwen1438
16Qwen3.8-27BQwen1437
17GLM-4.7Z.ai1435
18gemma-4-26B-A4B-itGoogle1434
19GLM-4.5Z.ai1430
20Mistral-Large-3-675B-Instruct-2512Mistral AI1429
21MiniMax-M3MiniMax1428
22DeepSeek-R1-0528DeepSeek1428
23MiMo-V2.5Xiaomi MiMo1427
24DeepSeek-V3.2-ExpDeepSeek1425
25Beam-1.0-501B-A23Breflection1424
26LongCat-Flash-ChatMeituan LongCat1423
27Mistral-Medium-3.5-128BMistral AI1421
28DeepSeek-V3.1DeepSeek1420
29DeepSeek-V3.1-TerminusDeepSeek1419
30Qwen3.5-122B-A10BQwen1417
31Qwen3-235B-A22B-Thinking-2507Qwen1415
32Inkling-SmallThinking Machines1414
33Mistral-Large-4-1T-A52BMistral AI1411
34MiMo-V2-FlashXiaomi MiMo1411
35Qwen3.5-27BQwen1409
36NVIDIA-Nemotron-3.5-Super-VL-120B-A12B-BF16NVIDIA1407
37MiniMax-M2.7MiniMax1405
38Inkling-2-SmallThinking Machines1403
39Step-3.5-FlashStepFun1402
40Hy3-previewTencent1401
41Qwen3-VL-235B-A22B-ThinkingQwen1401
42Qwen3.5-35B-A3BQwen1395
43Qwen3-235B-A22BQwen1394
44Muse-Glimmer-30Bmeta-models1389
45GLM-4.5-AirZ.ai1384
46NVIDIA-Nemotron-3-Super-120B-A12B-BF16NVIDIA1377
47GLM-4.6VZ.ai1377
48DeepSeek-R1DeepSeek1373
49Qwen3-Next-80B-A3B-ThinkingQwen1368
50gpt-oss-120bOpenAI1366
51granite-4.2-30bIBM Granite1362
52Ling-flash-2.0inclusionAI1361
53MiniMax-M2.5MiniMax1359
54INTELLECT-3Prime Intellect1356
55GLM-4.7-FlashZ.ai1351
56NVIDIA-Nemotron-3-Nano-30B-A3B-BF16NVIDIA1348
57MiniMax-M2MiniMax1340
58Qwen3-32BQwen1340
59Trinity-Large-ThinkingArcee AI1340
60GLM-4.5VZ.ai1333
61NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16NVIDIA1329
62QwQ-32BQwen1329
63granite-4.2-8bIBM Granite1319
64Qwen3-30B-A3BQwen1317
65Olmo-3-32B-ThinkAi21299
66granite-4.2-3bIBM Granite1292
67gpt-oss-20bOpenAI1288
68Olmo-3.1-32B-ThinkAi21271
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.