CG-AL-X271
Description
Per-model results
| Model | Attempt 1 | Attempt 2 | Avg score | Tokens / attempt | Runs |
|---|---|---|---|---|---|
| Claude Opus 5 | ✗ | ✓ | 81.3 / 100 | 15k / 16k | 3 |
| Claude Opus 5.5 | ✗ | ✓ | 81.3 / 100 | 14k / 6.1k | 3 |
| GPT-5.6 Terra | ✗ | ✓ | 81.3 / 100 | 8.5k / 4.7k | 3 |
| GPT-6 Sol | ✗ | ✓ | 81.3 / 100 | 8.0k / 4.1k | 3 |
| GPT-6.1 Sol | ✗ | ✓ | 81.3 / 100 | 8.0k / 3.1k | 3 |
| Gemini 3.8 Flash (OpenRouter) | ✗ | ✓ | 81.3 / 100 | 9.3k / 25k | 3 |
| GPT-5.6 Sol | ✗ | ✓ | 75.0 / 100 | 8.4k / 4.7k | 3 |
| GPT-6 Luna | ✗ | ✗ | 62.5 / 100 | 9.2k / 6.8k | 3 |
| GPT-5.6 Luna | ✗ | ✗ | 52.1 / 100 | 8.7k / 5.4k | 3 |
| Claude Sonnet 5 | ✗ | ✗ | 31.3 / 100 | 14k / 12k | 3 |
| Claude Haiku 4 5 | ✗ | ✗ | 20.8 / 100 | 12k / 3.5k | 3 |
| GLM 5.3 (OpenRouter) | ✗ | ✗ | 15.6 / 100 | 6.8k / 63k | 2 |
| MiniMax M3 | ✗ | ✗ | 10.4 / 100 | 3.1k / 14k | 3 |