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

31 models in coding ranked by LMArena score · updated Nov 11, 2026

#ModelScore
1GLM-5.3-FlashZ.ai1509
2Kimi-K3Moonshot AI1506
3DeepSeek-V4.1-FlashDeepSeek1504
4Kimi-K2.6Moonshot AI1489
5Kimi-K2.5Moonshot AI1484
6Kimi-K3.1-ProMoonshot AI1480
7Qwen3.8-27BQwen1479
8MiMo-V2.5Xiaomi MiMo1469
9MiniMax-M3MiniMax1468
10InklingThinking Machines1467
11Qwen3.5-397B-A17BQwen1463
12gemma-4-31B-itGoogle1459
13Mistral-Large-4-1T-A52BMistral AI1457
14Inkling-2-SmallThinking Machines1455
15Inkling-SmallThinking Machines1451
16gemma-4-26B-A4B-itGoogle1447
17Qwen3-VL-235B-A22B-InstructQwen1438
18Qwen3.5-122B-A10BQwen1435
19NVIDIA-Nemotron-3.5-Super-VL-120B-A12B-BF16NVIDIA1433
20Qwen3.5-27BQwen1427
21Qwen3-VL-235B-A22B-ThinkingQwen1425
22Muse-Glimmer-30Bmeta-models1418
23Qwen3.5-35B-A3BQwen1410
24GLM-4.6VZ.ai1391
25GLM-4.5VZ.ai1348
26gemma-3-27b-itGoogle1322
27Llama-4-Maverick-17B-128E-InstructMeta Llama1301
28Llama-4-Scout-17B-16E-InstructMeta Llama1286
29gemma-3-12b-itGoogle1281
30gemma-3n-E4B-itGoogle1268
31gemma-3-4b-itGoogle1230
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.