Best LLMs for Coding
Ranked by Arena coding rating — human preference votes from pairwise comparisons, not task-accuracy scores. LMArena dataset, CC BY 4.0.
| Rank | Model | Developer | Released | Arena coding rating | Blended $/1M |
|---|---|---|---|---|---|
| 1 | Claude Opus 5.5 | Anthropic | 2026-09-22 | 1551 | 8.00 |
| 2 | Claude Fable 5 | Anthropic | 2026-06-09 | 1550 | 20.00 |
| 3 | Gemini 4 ArgonNEW | 2026-10-13 | 1550 | 4.00 | |
| 4 | Claude Opus 4.6 | Anthropic | 2026-02-05 | 1546 | 10.00 |
| 5 | Claude Opus 4.7 | Anthropic | 2026-04-16 | 1546 | 10.00 |
| 6 | Muse WonderNEW | Meta | 2026-10-12 | 1545 | 8.00 |
| 7 | Muse Spark 1.2 | Meta | 2026-08-05 | 1543 | 2.00 |
| 8 | GPT-6 Astra | OpenAI | 2026-09-04 | 1542 | 20.00 |
| 9 | GPT-6.1 Sol | OpenAI | 2026-09-29 | 1541 | 4.00 |
| 10 | Kimi K3 | Moonshot AI | 2026-07-16 | 1540 | 6.00 |
| 11 | MiMo-V2.6-Pro | Xiaomi | 2026-09-22 | 1538 | 0.5437 |
| 12 | Claude Fable 5.1 | Anthropic | 2026-09-01 | 1537 | 20.00 |
| 13 | Claude Opus 5 | Anthropic | 2026-07-24 | 1534 | 10.00 |
| 14 | Muse Spark 1.3 | Meta | 2026-09-02 | 1534 | 2.00 |
| 15 | Muse Spark 1.1 | Meta | 2026-04-08 | 1533 | 2.00 |
| 16 | GPT-5.6 Sol | OpenAI | 2026-07-09 | 1533 | 8.00 |
| 17 | Gemini 3.8 Flash | 2026-09-02 | 1530 | 1.50 | |
| 18 | Claude Opus 4.8 | Anthropic | 2026-05-28 | 1529 | 10.00 |
| 19 | Claude Sonnet 5.5 | Anthropic | 2026-09-28 | 1529 | 4.00 |
| 20 | Claude Sonnet 4.6 | Anthropic | 2026-02-17 | 1528 | 6.00 |
| 21 | GLM-5.3-Flash | Zhipu AI | 2026-08-26 | 1526 | 0.1942 |
| 22 | Qwen3.8 Max | Alibaba | 2026-08-03 | 1525 | 3.00 |
| 23 | Claude Opus 4.5 | Anthropic | 2025-11-24 | 1522 | 10.00 |
| 24 | Gemini 3.1 Pro Preview | 2026-02-19 | 1521 | 4.50 | |
| 25 | MiMo-V2.5-Pro | Xiaomi | 2026-04-22 | 1521 | 0.5437 |
| 26 | GLM-5.3 | Zhipu AI | 2026-08-14 | 1520 | 1.9421 |
| 27 | Claude Sonnet 5 | Anthropic | 2026-06-30 | 1520 | 4.00 |
| 28 | Gemini 3.7 Flash | 2026-08-13 | 1520 | 1.50 | |
| 29 | GPT-6 Sol | OpenAI | 2026-09-22 | 1519 | 4.00 |
| 30 | Gemini 3.6 Flash | 2026-07-21 | 1518 | 1.50 | |
| 31 | GPT-5.6 Terra | OpenAI | 2026-07-09 | 1517 | 4.50 |
| 32 | Kimi K2.6 | Moonshot AI | 2026-04-20 | 1516 | 1.7125 |
| 33 | GLM-5.2 | Zhipu AI | 2026-06-13 | 1514 | 1.9421 |
| 34 | Grok 4.5 | xAI | 2026-07-08 | 1514 | 3.00 |
| 35 | GPT-5.5 Instant | OpenAI | 2026-05-05 | 1514 | 11.25 |
| 36 | GLM-5.1 | Zhipu AI | 2026-04-07 | 1513 | 1.5686 |
| 37 | Claude Sonnet 4.5 | Anthropic | 2025-09-29 | 1512 | 6.00 |
| 38 | Qwen3.6 Max Preview | Alibaba | 2026-04-20 | 1511 | 2.925 |
| 39 | GPT-5.4 | OpenAI | 2026-03-05 | 1511 | 5.625 |
| 40 | GPT-5.5 | OpenAI | 2026-04-23 | 1510 | 11.25 |
| 41 | MiMo-V2.6-Flash | Xiaomi | 2026-09-22 | 1510 | 0.175 |
| 42 | Grok 4.6 | xAI | 2026-08-12 | 1509 | 3.00 |
| 43 | Gemini 3.5 Flash | 2026-05-19 | 1508 | 3.375 | |
| 44 | GPT-6 Luna | OpenAI | 2026-09-22 | 1506 | 0.20 |
| 45 | Step 5 Preview | StepFun | 2026-09-16 | 1506 | 1.5318 |
| 46 | Beam 1.0 501B A23B | Reflection | 2026-10-05 | 1505 | — |
| 47 | Claude Opus 4.1 | Anthropic | 2025-08-05 | 1504 | — |
| 48 | Hy3 | Tencent | 2026-07-06 | 1503 | — |
| 49 | MiMo-V2-Pro | Xiaomi | 2026-03-18 | 1503 | — |
| 50 | Kimi K2.5 | Moonshot AI | 2026-01-27 | 1502 | 1.20 |
| 51 | Gemini 3.5 Flash Lite | 2026-07-21 | 1502 | 0.85 | |
| 52 | GPT-5.6 Luna | OpenAI | 2026-07-09 | 1502 | 0.45 |
| 53 | DeepSeek V4 Pro | DeepSeek | 2026-04-24 | 1500 | 1.98 |
| 54 | Qwen3.7 Plus | Alibaba | 2026-06-02 | 1500 | 0.70 |
| 55 | Qwen3.8 27B | Alibaba | 2026-08-14 | 1498 | 1.125 |
| 56 | GLM-5 | Zhipu AI | 2026-02-12 | 1498 | 1.1205 |
| 57 | Mistral Large 4 | Mistral | 2026-10-06 | 1496 | 1.05 |
| 58 | Qwen3.6 Plus | Alibaba | 2026-04-02 | 1496 | 1.125 |
| 59 | MiniMax-M3 | MiniMax | 2026-06-01 | 1495 | 0.525 |
| 60 | GPT-5.4 mini | OpenAI | 2026-03-17 | 1495 | 1.6875 |
| 61 | Grok 4.7 | xAI | 2026-09-21 | 1494 | 3.00 |
| 62 | Inkling | Thinking Machines | 2026-07-15 | 1494 | — |
| 63 | Qwen3.5 397B-A17B | Alibaba | 2026-02-15 | 1491 | 1.35 |
| 64 | MiMo-V2.5 | Xiaomi | 2026-04-22 | 1491 | 0.175 |
| 65 | Claude Haiku 5.5NEW | Anthropic | 2026-10-07 | 1489 | 0.20 |
| 66 | MiMo-V2-Omni | Xiaomi | 2026-03-18 | 1488 | — |
| 67 | GLM-5V-Turbo | Zhipu AI | 2026-04-01 | 1488 | 1.3819 |
| 68 | Grok 4.3 | xAI | 2026-04-17 | 1488 | 1.5625 |
| 69 | Kimi K2 Thinking Turbo | Moonshot AI | 2025-11-06 | 1486 | — |
| 70 | GLM-4.7 | Zhipu AI | 2025-12-22 | 1484 | 0.859 |
| 71 | DeepSeek V4 Flash | DeepSeek | 2026-04-24 | 1484 | 0.525 |
| 72 | GPT-5.2 | OpenAI | 2025-12-11 | 1482 | 4.8125 |
| 73 | Claude Haiku 4.5 | Anthropic | 2025-10-15 | 1482 | 2.00 |
| 74 | Qwen3 Max Preview | Alibaba | 2025-09-05 | 1480 | — |
| 75 | MiniMax-M2.7 | MiniMax | 2026-03-18 | 1480 | 0.525 |
| 76 | Muse Glimmer 30B | Meta | 2026-08-10 | 1478 | — |
| 77 | Mistral Medium 3.5 | Mistral | 2026-04-30 | 1477 | 3.00 |
| 78 | GPT-5.1 | OpenAI | 2025-11-13 | 1476 | 3.4375 |
| 79 | Inkling Small | Thinking Machines | 2026-07-30 | 1472 | — |
| 80 | Qwen3 235B-A22B Instruct 2507 | Alibaba | 2025-07-21 | 1472 | — |
| 81 | GPT-5 | OpenAI | 2025-08-07 | 1471 | 3.4375 |
| 82 | Mistral Large 3 | Mistral | 2025-12-02 | 1471 | 0.75 |
| 83 | DeepSeek V3.2 | DeepSeek | 2025-12-01 | 1469 | — |
| 84 | Claude Opus 4 | Anthropic | 2025-05-22 | 1465 | — |
| 85 | Gemini 2.5 Pro | 2025-06-17 | 1464 | 3.4375 | |
| 86 | DeepSeek R1 0528 | DeepSeek | 2025-05-28 | 1464 | — |
| 87 | Qwen3 VL 235B A22B Instruct | Alibaba | 2025-09-23 | 1463 | 0.70 |
| 88 | Solar Pro 4 | Upstage | 2026-08-06 | 1461 | 0.525 |
| 89 | Grok 4.1 Fast (Reasoning) | xAI | 2025-11-19 | 1461 | — |
| 90 | GPT-5.4 nano | OpenAI | 2026-03-17 | 1460 | 0.4625 |
| 91 | Qwen3.5 122B-A10B | Alibaba | 2026-02-23 | 1458 | 1.10 |
| 92 | GLM-4.6 | Zhipu AI | 2025-09-30 | 1458 | 1.00 |
| 93 | Gemini 3.1 Flash Lite Preview | 2026-03-03 | 1458 | 0.5625 | |
| 94 | Qwen3-Coder 480B-A35B Instruct | Alibaba | 2025-07-22 | 1457 | 3.00 |
| 95 | X-Ai/Grok-4-Fast-Non-Reasoning | xAI | 2025-12-18 | 1456 | — |
| 96 | GLM-4.5 | Zhipu AI | 2025-07-28 | 1455 | 1.00 |
| 97 | Hy3 preview | Tencent | 2026-04-20 | 1452 | — |
| 98 | Qwen3 VL 235B A22B Thinking | Alibaba | 2025-09-23 | 1452 | 1.30 |
| 99 | Qwen3.5 27B | Alibaba | 2026-02-23 | 1451 | 0.825 |
| 100 | Step 3.5 Flash | StepFun | 2026-01-29 | 1449 | 0.1569 |
| 101 | Claude Sonnet 4 | Anthropic | 2025-05-22 | 1449 | — |
| 102 | DeepSeek V3.1 | DeepSeek | 2025-08-21 | 1447 | — |
| 103 | MiniMax-M2.5 | MiniMax | 2026-02-12 | 1446 | 0.525 |
| 104 | Qwen3-Next 80B-A3B Instruct | Alibaba | 2025-09 | 1444 | 0.875 |
| 105 | DeepSeek-R1 | DeepSeek | 2025-01-20 | 1444 | — |
| 106 | Trinity Large Preview | Arcee AI | 2026-01-27 | 1443 | — |
| 107 | Qwen3 235B-A22B Thinking 2507 | Alibaba | 2025-07-25 | 1442 | — |
| 108 | X-Ai/Grok-4-Fast-Reasoning | xAI | 2025-12-18 | 1439 | — |
| 109 | Qwen3 30B-A3B Instruct 2507 | Alibaba | 2025-07-28 | 1437 | — |
| 110 | Claude Sonnet 3.5 v2 | Anthropic | 2024-10-22 | 1436 | — |
| 111 | Qwen3.5 Flash | Alibaba | 2026-02-23 | 1435 | 0.175 |
| 112 | Qwen3.5 35B-A3B | Alibaba | 2026-02-23 | 1435 | 0.6875 |
| 113 | MiMo-V2-Flash | Xiaomi | 2025-12-16 | 1432 | — |
| 114 | GPT-5 Mini | OpenAI | 2025-08-07 | 1432 | 0.6875 |
| 115 | Qwen3 235B-A22B | Alibaba | 2025-04-28 | 1431 | 1.225 |
| 116 | Claude Sonnet 3.7 | Anthropic | 2025-02-19 | 1431 | — |
| 117 | DeepSeek V3 0324 | DeepSeek | 2025-03-24 | 1428 | — |
| 118 | GLM-4.5-Air | Zhipu AI | 2025-07-28 | 1426 | 0.3137 |
| 119 | Gemini 2.5 Flash | 2025-06-17 | 1424 | 0.85 | |
| 120 | GLM-4.7-Flash | Zhipu AI | 2026-01-19 | 1423 | — |
| 121 | Qwen3-Next 80B-A3B (Thinking) | Alibaba | 2025-09 | 1421 | 1.875 |
| 122 | o3-mini | OpenAI | 2024-12-20 | 1416 | 1.925 |
| 123 | GLM-4.6V | Zhipu AI | 2025-12-08 | 1416 | 0.2241 |
| 124 | Trinity Large Thinking | Arcee AI | 2026-04-01 | 1413 | — |
| 125 | Qwen3 32B | Alibaba | 2025-04-28 | 1406 | 1.225 |
| 126 | Llama 3.3 Nemotron Super 49B v1.5 | Nvidia | 2025-07-25 | 1403 | — |
| 127 | GLM-4.5V | Zhipu AI | 2025-08-11 | 1402 | 0.4482 |
| 128 | Mercury 2 | Inception | 2026-02-24 | 1392 | 0.375 |
| 129 | GPT OSS 120B | OpenAI | 2025-08-05 | 1389 | — |
| 130 | Command A | Cohere | 2025-03-13 | 1388 | 4.375 |
| 131 | DeepSeek-V3 | DeepSeek | 2024-12-26 | 1387 | — |
| 132 | Claude Haiku 3.5 | Anthropic | 2024-10-22 | 1385 | — |
| 133 | GPT-5 Nano | OpenAI | 2025-08-07 | 1385 | 0.1375 |
| 134 | Qwen3 30B A3B | Alibaba | 2025-04-28 | 1385 | — |
| 135 | MiniMax-M2 | MiniMax | 2025-10-27 | 1383 | 0.525 |
| 136 | QwQ 32B | Alibaba | 2025-03-05 | 1383 | — |
| 137 | Granite 4.2 8B | IBM | 2026-08-25 | 1370 | — |
| 138 | GPT-4o (2024-05-13) | OpenAI | 2024-05-13 | 1369 | 7.50 |
| 139 | GPT OSS 20B | OpenAI | 2025-08-05 | 1368 | — |
| 140 | Llama 3.3 Nemotron Super 49B v1 | Nvidia | 2025-03-18 | 1362 | — |
| 141 | Mistral Small 3.1 24B | Mistral | 2025-03-17 | 1361 | — |
| 142 | GPT-4o (2024-08-06) | OpenAI | 2024-08-06 | 1359 | 4.375 |
| 143 | Gemma 3 27B IT | 2025-03-12 | 1357 | — | |
| 144 | Mistral Large 2.1 | Mistral | 2024-11-18 | 1346 | — |
| 145 | Llama-3.3-70B-Instruct | Meta | 2024-12-06 | 1346 | — |
| 146 | Qwen2.5-Coder-32B-Instruct | Alibaba | 2024-11-12 | 1342 | — |
| 147 | Llama-3.1-70B-Instruct | Meta | 2024-07-23 | 1333 | — |
| 148 | Granite-4.0-H-Small | IBM | 2025-10-02 | 1330 | — |
| 149 | Llama 3.1 Nemotron 70B Instruct | Nvidia | 2025-04-15 | 1328 | — |
| 150 | Gemma 3 12B IT | 2025-03-12 | 1316 | — | |
| 151 | Mistral Small 3 | Mistral | 2025-01-30 | 1312 | — |
| 152 | Gemma 3n E4B IT | 2025-06-26 | 1307 | — | |
| 153 | Gemma 2 27B | 2024-07-13 | 1304 | — | |
| 154 | Claude Haiku 3 | Anthropic | 2024-03-13 | 1301 | — |
| 155 | Aya Expanse 32B | Cohere | 2024-10-24 | 1292 | 0.75 |
| 156 | Command R | Cohere | 2024-08-30 | 1281 | 0.2625 |
| 157 | Gemma 3 4B IT | 2025-03-12 | 1273 | — | |
| 158 | Llama-3.1-8B-Instruct | Meta | 2024-07-23 | 1259 | — |
| 159 | Aya Expanse 8B | Cohere | 2024-10-24 | 1255 | — |
| 160 | Mixtral 8x7B Instruct v0.1 | Mistral | 2023-12-11 | 1240 | — |
| 161 | Gemma 2 2B IT | 2024-07-31 | 1191 | — | |
| 162 | SOLAR 10.7B Instruct v1.0 | Upstage | 2023-12-12 | 1184 | — |
| 163 | Llama 3.2 3B Instruct | Meta | 2024-09-25 | 1176 | — |
| 164 | Llama 3.2 1B Instruct | Meta | 2024-09-25 | 1149 | — |
FAQ
How is the Arena coding rating measured?
It's the model's Arena coding rating, one canonical row per model. These are human preference votes from pairwise comparisons, not task-accuracy scores.
What is the top-ranked model for coding right now?
Currently Claude Opus 5.5 by Anthropic, at 1551 on this metric.
How is this different from the main leaderboard?
The main /leaderboard/ page ranks by Epoch's ECI. This page ranks only by arena coding rating — useful when your workload is specifically about that skill.
Why do some models not appear here?
Only models with a published arena coding rating are ranked — 677 of 841 tracked models have no published score here and are excluded rather than shown with a misleading zero.