Benchmarks
WebDev Arena
Models scored
82
Best score
1653.9
Metric
Arena Score
WebDev Arena is an AI evaluation tracked by Epoch AI, with 82 scored model versions on record. The scores here are reported by third parties rather than produced by Epoch AI running the evaluation itself.
Scored models span June 2025 to June 2026. The best score recorded is 1653.93, measured as Arena Score. Models come from Anthropic, Z.ai (Zhipu AI), Alibaba, MiniMax, Moonshot, Meta AI, Google, OpenAI and others.
Full record
- Score metric
- Arena Score
- Models scored
- 82
- Best score recorded
- 1653.9
- Earliest model scored
- Jun 17, 2025
- Most recent model scored
- Jun 16, 2026
- Third-party reported
- Yes
Leaderboard
| 01 | claude-fable-5_unknown | Anthropic | 1653.9 |
| 02 | glm-5.2_max | Z.ai (Zhipu AI) | 1593.3 |
| 03 | claude-opus-4-7_unknown | Anthropic | 1566.8 |
| 04 | claude-opus-4-7 | Anthropic | 1562.4 |
| 05 | claude-opus-4-6_unknown | Anthropic | 1556.3 |
| 06 | claude-opus-4-6 | Anthropic | 1555.3 |
| 07 | claude-opus-4-8_unknown | Anthropic | 1552.2 |
| 08 | claude-opus-4-8_none | Anthropic | 1545 |
| 09 | qwen3.7-max | Alibaba | 1540.8 |
| 10 | glm-5.1 | Z.ai (Zhipu AI) | 1534 |
| 11 | MiniMax-M3 | MiniMax | 1527.8 |
| 12 | kimi-k2.6 | Moonshot | 1526 |
| 13 | claude-sonnet-4-6 | Anthropic | 1522.9 |
| 14 | muse-spark | Meta AI | 1512.5 |
| 15 | claude-opus-4-5-20251101_32K | Anthropic | 1512 |
| 16 | gemini-3.5-flash_unknown | 1506.4 | |
| 17 | gpt-5.5_xhigh | OpenAI | 1504.7 |
| 18 | qwen3.6-max-preview | Alibaba | 1485.6 |
| 19 | gpt-5.2-2025-12-11_high | OpenAI | 1480 |
| 20 | kimi-k2.7-code | Moonshot | 1479.2 |
| 21 | claude-opus-4-5-20251101 | Anthropic | 1479 |
| 22 | gpt-5.5_high | OpenAI | 1478.9 |
| 23 | mimo-v2.5-pro | Xiaomi Corp | 1471.3 |
| 24 | gemini-3-pro-preview | Google DeepMind | 1471 |
| 25 | deepseek-v4-pro_unknown | DeepSeek | 1463.6 |
| 26 | gemini-3.1-pro-preview | Google DeepMind | 1461.5 |
| 27 | qwen3.6-plus | Alibaba | 1460.4 |
| 28 | gpt-5.4-2026-03-05_high | OpenAI | 1457.2 |
| 29 | gemini-3-flash-preview | Google DeepMind | 1454 |
| 30 | glm-5 | Z.ai (Zhipu AI) | 1447.3 |
| 31 | gpt-5.5_unknown | OpenAI | 1444.3 |
| 32 | glm-4.7 | Z.ai (Zhipu AI) | 1441 |
| 33 | gpt-5.4-2026-03-05_medium | OpenAI | 1437.3 |
| 34 | mimo-v2-pro | Xiaomi Corp | 1432.2 |
| 35 | mimo-v2.5 | Xiaomi Corp | 1432.1 |
| 36 | kimi-k2.5 | Moonshot | 1430.6 |
| 37 | kimi-k2.5_none | Moonshot | 1408 |
| 38 | gpt-5.3-codex | OpenAI | 1406.7 |
| 39 | gpt-5.2-2025-12-11_unknown | OpenAI | 1404.4 |
| 40 | gpt-5.4-mini-2026-03-17_high | OpenAI | 1402.2 |
| 41 | MiniMax-M2.7 | MiniMax | 1400.5 |
| 42 | grok-4-20 | xAI | 1395.2 |
| 43 | gpt-5-2025-08-07_medium | OpenAI | 1395 |
| 44 | gpt-5.2-2025-12-11_medium | OpenAI | 1394 |
| 45 | qwen3.5-plus | Alibaba | 1393.2 |
| 46 | MiniMax-M2.1 | MiniMax | 1391.6 |
| 47 | claude-sonnet-4-5-20250929_32K | Anthropic | 1391 |
| 48 | gpt-5.4-2026-03-05_unknown | OpenAI | 1388.3 |
| 49 | gpt-5.1-2025-11-13_medium | OpenAI | 1387 |
| 50 | claude-sonnet-4-5-20250929 | Anthropic | 1387 |
| 51 | claude-opus-4-1-20250805 | Anthropic | 1386 |
| 52 | claude-opus-4-1-20250805_unknown | Anthropic | 1385.8 |
| 53 | MiniMax-M2.5 | MiniMax | 1382.1 |
| 54 | gemma-4-31b-it | Google DeepMind | 1379.5 |
| 55 | grok-4-3 | xAI | 1376.6 |
| 56 | DeepSeek-V3.2-Exp_thinking | DeepSeek | 1367 |
| 57 | glm-4.6 | Z.ai (Zhipu AI),Tsinghua University | 1366 |
| 58 | gemma-4-26b-a4b | Google DeepMind | 1359.3 |
| 59 | gpt-5.1-2025-11-13_low | OpenAI | 1359 |
| 60 | qwen3.5-27B | Alibaba | 1356.9 |
More benchmarks
SourceEpoch AI, 'AI Benchmarking Hub'. Published online at epoch.ai. Retrieved 2026-07-29 from https://epoch.ai/benchmarks. Licensed under CC BY 4.0.
← All benchmarks