Atomic Task Success on CrafText Atom (test)
56Success RateSI_pixel
Evaluation Results
| Method | Links | |
|---|---|---|
| SI_pixelTraining Iteration=Cycle 12026.04 | 56 | |
| Claude Sonnet 4.62026.04 | 24 | |
| Qwen/QWQ-32B2026.04 | 22 | |
| OpenAI GPT-4.12026.04 | 14 | |
| Gemma-27B2026.04 | 6 |
| Method | Links | |
|---|---|---|
| SI_pixelTraining Iteration=Cycle 12026.04 | 56 | |
| Claude Sonnet 4.62026.04 | 24 | |
| Qwen/QWQ-32B2026.04 | 22 | |
| OpenAI GPT-4.12026.04 | 14 | |
| Gemma-27B2026.04 | 6 |