Code Generation on LiveCodeBench (Acceptance length, Delta %)
Evaluation results
| Method | Links | | |
|---|
| TTSTarget Model=Qwen/Qwen3-4B2026.05 | | 4.7 | 55.2 |
|---|
| TTSTarget Model=Qwen/Qwen3-8B2026.05 | | 4.5 | 55.2 |
|---|
| TTSTarget Model=Qwen/Qwen3.5-122B2026.05 | | 4 | 63.5 |
|---|
| DFlashTarget Model=Qwen/Qwen3-4B2026.05 | | 3 | — |
|---|
| DFlashTarget Model=Qwen/Qwen3-8B2026.05 | | 2.9 | — |
|---|
| DFlashTarget Model=Qwen/Qwen3.5-122B2026.05 | | 2.4 | — |
|---|