CG-AL-X254
Description
Per-model results
| Model | Attempt 1 | Attempt 2 | Avg score | Tokens / attempt | Runs |
|---|---|---|---|---|---|
| Claude Opus 5.5 | ✓ | — | 100.0 / 100 | 10k / 6.3k | 3 |
| Gemini 3.8 Flash (OpenRouter) | ✓ | — | 100.0 / 100 | 6.5k / 29k | 3 |
| GPT-6.1 Sol | ✓ | ✓ | 90.6 / 100 | 6.1k / 2.2k | 3 |
| Claude Opus 5 | ✗ | ✓ | 81.3 / 100 | 11k / 6.5k | 3 |
| Claude Sonnet 5 | ✗ | ✓ | 81.3 / 100 | 11k / 10k | 3 |
| GLM 5.3 (OpenRouter) | ✗ | ✓ | 81.3 / 100 | 6.6k / 45k | 2 |
| GPT-5.6 Sol | ✗ | ✗ | 62.5 / 100 | 7.4k / 4.1k | 3 |
| GPT-5.6 Terra | ✗ | ✗ | 62.5 / 100 | 7.3k / 3.2k | 3 |
| GPT-6 Sol | ✗ | ✗ | 62.5 / 100 | 6.8k / 2.8k | 3 |
| MiniMax M3 | ✗ | ✓ | 54.2 / 100 | 5.2k / 10k | 3 |
| Claude Haiku 4 5 | ✗ | ✗ | 52.1 / 100 | 12k / 1.6k | 3 |
| GPT-6 Luna | ✗ | ✗ | 41.7 / 100 | 6.3k / 3.5k | 3 |
| GPT-5.6 Luna | ✗ | ✗ | 31.3 / 100 | 6.8k / 2.7k | 3 |