CG-AL-X289
Description
Per-model results
| Model | Attempt 1 | Attempt 2 | Avg score | Tokens / attempt | Runs |
|---|---|---|---|---|---|
| Claude Opus 5.5 | ✓ | — | 100.0 / 100 | 8.8k / 9.3k | 3 |
| GPT-6.1 Sol | ✓ | — | 100.0 / 100 | 5.0k / 4.9k | 3 |
| Gemini 3.8 Flash (OpenRouter) | ✓ | ✓ | 90.6 / 100 | 6.0k / 38k | 3 |
| Claude Opus 5 | ✗ | ✓ | 81.3 / 100 | 10k / 19k | 3 |
| Claude Sonnet 5 | ✗ | ✓ | 81.3 / 100 | 10k / 14k | 3 |
| GPT-5.6 Sol | ✗ | ✓ | 64.6 / 100 | 5.8k / 4.0k | 3 |
| GPT-5.6 Terra | ✗ | ✓ | 64.6 / 100 | 5.8k / 3.5k | 3 |
| Claude Haiku 4 5 | ✗ | ✗ | 62.5 / 100 | 9.7k / 2.9k | 3 |
| GPT-5.6 Luna | ✗ | ✓ | 58.3 / 100 | 7.4k / 4.0k | 3 |
| GPT-6 Luna | ✗ | ✓ | 47.9 / 100 | 5.6k / 5.3k | 3 |
| GPT-6 Sol | ✗ | ✗ | 20.8 / 100 | 5.2k / 3.1k | 3 |
| MiniMax M3 | ✗ | ✗ | 0.0 / 100 | 491 / 16k | 3 |
| GLM 5.3 (OpenRouter) | ✗ | ✗ | 0.0 / 100 | 3.0k / 64k | 2 |