General Capability on Aggregated 7-Metric Suite (test)
43.9Average ScoreG-Zero
Evaluation Results
| Method | Links | |
|---|---|---|
| G-ZeroBackbone=Llama-3.1-8B-Instruct, Training Method=G-Zero, Training Round=22026.05 | 43.9 | |
| G-ZeroBackbone=Llama-3.1-8B-Instruct, Training Method=G-Zero, Training Round=12026.05 | 43.08 | |
| base modelBackbone=Llama-3.1-8B-Instruct, Training Method=Base2026.05 | 42.77 | |
| R-ZeroBackbone=Llama-3.1-8B-Instruct, Training Method=R-Zero2026.05 | 40.89 | |
| G-ZeroBackbone=Qwen3-8B-Base, Training Method=G-Zero, Training Round=22026.05 | 35.43 | |
| G-ZeroBackbone=Qwen3-8B-Base, Training Method=G-Zero, Training Round=12026.05 | 34.96 | |
| base modelBackbone=Qwen3-8B-Base, Training Method=Base2026.05 | 33.95 | |
| R-ZeroBackbone=Qwen3-8B-Base, Training Method=R-Zero2026.05 | 32.59 |