123 B
123 B
license
| license |
|---|
| other |
ABForge-Qwen3-8B-Combined-ckpt200
Combined SFT+RL ckpt sweep: step 200. bench_44: task1 C=56.4.
| license |
|---|
| other |
Combined SFT+RL ckpt sweep: step 200. bench_44: task1 C=56.4.