Arena Code WebDev — Gaming
Code
Arena Code WebDev rating for interactive browser games. Anonymous generated games are compared through blind pairwise user votes.
Top models (higher is better)
| Model | Score |
|---|---|
| Claude Opus 5 | 1801 |
| Kimi K3 | 1716 |
| Claude Fable 5 | 1705 |
| GLM-5.2 | 1621 |
| Opus 4.8 | 1617 |
| DeepSeek-V4-Flash | 1601 |
| Opus 4.7 | 1595 |
| Grok 4.5 | 1583 |
| Opus 4.6 | 1582 |
| Sonnet 5 | 1579 |
| Gemini 3.6 Flash | 1551 |
| GLM-5.1 | 1547 |
| Hy3 | 1545 |
| Sonnet 4.6 | 1544 |
| Muse Spark 1.1 | 1543 |
| Gemini 3.5 Flash | 1521 |
| Kimi K2.6 | 1499 |
| MiMo-V2.5-Pro | 1495 |
| Opus 4.5 | 1482 |
| Kimi K2.7 Code | 1478 |
| MiniMax M3 | 1477 |
| Qwen3.6 Plus (2026-04-02) | 1468 |
| Gemini 3.1 Pro Preview | 1467 |
| Gemini 3 Flash Preview | 1454 |
| DeepSeek-V4-Pro | 1453 |
| Gemini 3 Pro Preview | 1453 |
| GPT-5.3-Codex | 1446 |
| MiMo-V2-Pro | 1436 |
| GPT-5.2 | 1435 |
| GLM-5 | 1434 |
| GPT-5 | 1432 |
| MiMo-V2.5 | 1427 |
| GLM-4.7 | 1427 |
| Kimi K2.5 | 1425 |
| GPT-5.4 Mini | 1424 |
| Opus 4.1 | 1424 |
| GLM 5V Turbo | 1407 |
| GPT-5.1 | 1401 |
| Qwen3.5 397B A17B | 1400 |
| GPT-5.4 | 1398 |