CG-AL-X187
Description
Per-model results
| Model | Attempt 1 | Attempt 2 | Avg score | Tokens / attempt | Runs |
|---|---|---|---|---|---|
| Claude Opus 5 | ✗ | ✓ | 81.3 / 100 | 6.5k / 19k | 3 |
| GPT-6 Luna | ✗ | ✓ | 81.3 / 100 | 3.7k / 2.7k | 3 |
| GPT-5.6 Luna | ✗ | ✓ | 75.0 / 100 | 3.7k / 3.1k | 3 |
| GPT-5.6 Sol | ✗ | ✓ | 75.0 / 100 | 3.8k / 2.5k | 3 |
| Gemini 3.8 Flash (OpenRouter) | ✗ | ✓ | 68.8 / 100 | 4.4k / 22k | 3 |
| Claude Opus 5.5 | ✗ | ✓ | 64.6 / 100 | 6.1k / 2.7k | 3 |
| GPT-5.6 Terra | ✗ | ✓ | 64.6 / 100 | 3.6k / 2.3k | 3 |
| GPT-6.1 Sol | ✗ | ✓ | 64.6 / 100 | 3.5k / 1.3k | 3 |
| GPT-6 Sol | ✗ | ✗ | 62.5 / 100 | 3.7k / 1.8k | 3 |
| Claude Haiku 4 5 | ✗ | ✗ | 52.1 / 100 | 5.0k / 1.5k | 3 |
| MiniMax M3 | ✗ | ✗ | 41.7 / 100 | 4.2k / 12k | 3 |
| Claude Sonnet 5 | ✗ | ✗ | 31.3 / 100 | 5.6k / 3.8k | 3 |
| GLM 5.3 (OpenRouter) | ✗ | ✗ | 31.3 / 100 | 3.9k / 51k | 2 |