Coding on LiveCodeBench v1-5
28.3AccuracyAZR
Evaluation Results
| Method | Links | |
|---|---|---|
| AZRModel Backbone=Qwen3-8B-Base2026.06 | 28.3 | |
| INFUSERModel Backbone=Qwen3-8B-Base2026.06 | 28.01 | |
| R-ZeroModel Backbone=Qwen3-8B-Base2026.06 | 25.91 | |
| BaseModel Backbone=Qwen3-8B-Base2026.06 | 25.23 | |
| INFUSERModel Backbone=Qwen3-4B-Base2026.06 | 22.35 | |
| AZRModel Backbone=Qwen3-4B-Base2026.06 | 22.33 | |
| R-ZeroModel Backbone=Qwen3-4B-Base2026.06 | 21.82 | |
| BaseModel Backbone=Qwen3-4B-Base2026.06 | 20.68 |