Reasoning / Reward on RR
73.2RR AverageQwen2.5-Coder-14B-Ins
Evaluation Results
| Method | Links | |
|---|---|---|
| Qwen2.5-Coder-14B-InsModel Family=AZR Series, Variant=Original2026.06 | 73.2 | |
| ANCHOR-8BModel Family=R-Zero Series, Variant=+R-Zero w/ ANCHOR2026.06 | 73.2 | |
| ANCHOR-Coder-14BModel Family=AZR Series, Variant=+AZR w/ ANCHOR2026.06 | 70.7 | |
| AZR-Coder-14BModel Family=AZR Series, Variant=+AZR2026.06 | 65.9 | |
| ANCHOR-Coder-7BModel Family=AZR Series, Variant=+AZR w/ ANCHOR2026.06 | 61 | |
| Qwen3-8B-BaseModel Family=R-Zero Series, Variant=Original2026.06 | 61 | |
| AZR-Coder-3BModel Family=AZR Series, Variant=+AZR2026.06 | 58.5 | |
| ANCHOR-Coder-3BModel Family=AZR Series, Variant=+AZR w/ ANCHOR2026.06 | 58.5 | |
| Qwen2.5-Coder-7B-InsModel Family=AZR Series, Variant=Original2026.06 | 58.5 | |
| Qwen2.5-7B-InsModel Family=AZR Series, Variant=Original2026.06 | 56.1 | |
| Qwen3-4B-BaseModel Family=R-Zero Series, Variant=Original2026.06 | 56.1 | |
| Qwen2.5-Coder-3B-InsModel Family=AZR Series, Variant=Original2026.06 | 53.7 | |
| AZR-Coder-7BModel Family=AZR Series, Variant=+AZR2026.06 | 53.7 | |
| ANCHOR-4BModel Family=R-Zero Series, Variant=+R-Zero w/ ANCHOR2026.06 | 53.7 | |
| ANCHOR-7BModel Family=AZR Series, Variant=+AZR w/ ANCHOR2026.06 | 46.3 | |
| AZR-7BModel Family=AZR Series, Variant=+AZR2026.06 | 36.6 | |
| R-Zero-8BModel Family=R-Zero Series, Variant=+R-Zero2026.06 | 34.1 | |
| R-Zero-4BModel Family=R-Zero Series, Variant=+R-Zero2026.06 | 29.3 |