| model \ task | long-context-needle | py-fix-bug | py-refactor |
|---|---|---|---|
| qwen3.6-27b | — | 5/5 · 100% | 5/5 · 100% |
| gemma-4-31b | — | 5/5 · 100% | 5/5 · 100% |
| qwen3.6-35b-a3b | — | 5/5 · 100% | 5/5 · 100% |
| ornith-1.0-35b | — | 5/5 · 100% | 5/5 · 100% |
| north-mini-code-1.0 | 4/5 · 93% | 5/5 · 100% | 5/5 · 100% |
| bielik-11b-v3 | 5/5 · 100% | 0/5 · 60% | 0/5 · 0% |
| gemma-4-26b-a4b | 5/5 · 100% | 5/5 · 100% | 5/5 · 100% |
Cells show passes / repetitions. Click a cell for the produced code, transcript and verify output.