VoxelBench Text Rating
Games · 2025-09-12
VoxelBench text rating is the Glicko-2 leaderboard track for models creating voxel builds from text prompts. Ratings come from head-to-head human votes over rendered voxel constructions.
Top models (higher is better)
| Model | Score |
|---|---|
| GPT-6.1 Sol | 2622 |
| GPT-6 Astra | 2611 |
| Claude Opus 5.5 | 2575 |
| GPT-6 Sol | 2398 |
| Claude Opus 5 | 2266 |
| Kimi K3.1 Pro | 2265 |
| GPT-5.6 Sol | 2235 |
| Grok 4.6 | 2068 |
| Claude Fable 5 | 2038 |
| DeepSeek V4.1 Flash 1017 | 2024 |
| GPT-5.5 Pro | 1943 |
| Kimi K3 | 1927 |
| GPT-5.6 Terra | 1908 |
| GPT-5.5 | 1898 |
| Muse Wonder | 1857 |
| DeepSeek V4.1 Flash | 1830 |
| GPT-6 Luna | 1829 |
| Gemini 3.7 Flash | 1822 |
| GPT-5.6 Luna | 1821 |
| Gemini 3.8 Flash | 1797 |
| Gemini 3.6 Flash | 1791 |
| MiniMax M3.1 Flash | 1784 |
| Grok 4.5 | 1771 |
| GLM-5.3 | 1768 |
| Claude Haiku 5.5 | 1739 |
| Muse Spark 1.3 | 1678 |
| Gemini 3.1 Pro Preview | 1666 |
| Gemini 3.5 Flash | 1655 |
| Qwen3.7-Max (source unspecified) | 1595 |
| Opus 4.8 | 1590 |
| DeepSeek V4 Pro 0813 | 1587 |
| GLM-5.2 | 1555 |
| Sonnet 5 | 1537 |
| Gemini 2.5 Deep Think | 1494 |
| Muse Spark 1.1 | 1489 |
| Gemini 3 Deep Think (source unspecified) | 1468 |
| GPT-5.2 | 1440 |
| GLM-5.1 | 1415 |
| GPT-5.4 | 1412 |
| Opus 4.6 | 1396 |