Arena Code WebDev — HTML
Code
Arena Code WebDev rating for websites generated as HTML. Anonymous outputs are compared through blind pairwise user votes.
Top models (higher is better)
| Model | Score |
|---|---|
| Claude Opus 5 | 1685 |
| Claude Fable 5 | 1676 |
| Kimi K3 | 1655 |
| Grok 4.5 | 1588 |
| Opus 4.7 | 1563 |
| Opus 4.8 | 1560 |
| Opus 4.6 | 1559 |
| GLM-5.2 | 1544 |
| Gemini 3.6 Flash | 1543 |
| GLM-5.1 | 1533 |
| Muse Spark 1.1 | 1532 |
| Gemini 3.5 Flash | 1529 |
| DeepSeek-V4-Flash | 1527 |
| Sonnet 4.6 | 1526 |
| Sonnet 5 | 1521 |
| Gemini 3.1 Pro Preview | 1498 |
| Kimi K2.6 | 1496 |
| MiMo-V2.5-Pro | 1486 |
| Opus 4.5 | 1481 |
| Gemini 3 Pro Preview | 1469 |
| Qwen3.6 Plus (2026-04-02) | 1464 |
| MiniMax M3 | 1464 |
| Kimi K2.7 Code | 1459 |
| GLM-5 | 1453 |
| GLM-4.7 | 1446 |
| Gemini 3 Flash Preview | 1444 |
| MiMo-V2-Pro | 1440 |
| DeepSeek-V4-Pro | 1438 |
| Kimi K2.5 | 1431 |
| GPT-5.2 | 1430 |
| Hy3 | 1429 |
| GPT-5 | 1429 |
| GPT-5.4 | 1429 |
| GPT-5.4 Mini | 1428 |
| MiMo-V2.5 | 1423 |
| GPT-5.3-Codex | 1420 |
| MiniMax M2.5 | 1415 |
| Qwen3.5 397B A17B | 1414 |
| MiniMax M2.7 | 1404 |
| GPT-5.1 | 1401 |