Role-playing agent evaluation on LLM Court 5 legal scenarios 1.0 (test)
92.5QS d BRF ScoreLlama-3.1-8B
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| Llama-3.1-8BQuantization=4-bit2026.04 | 92.5 | 3 | |
| Gemma-2-9B-itQuantization=4-bit2026.04 | 89.3 | 0 | |
| Qwen-3-8BQuantization=4-bit2026.04 | 89.3 | 1 | |
| Gemma-3-12B-itQuantization=4-bit2026.04 | 82.3 | 0 | |
| Hermes-3-8BQuantization=5-bit2026.04 | 81 | 3 | |
| Phi-4-14BQuantization=4-bit2026.04 | 76 | 0 | |
| Qwen-3-14BQuantization=4-bit2026.04 | 73 | 0 |