Red-teaming on AI Supremacy principle v1 (test)
11.7Mean Best Category ScoreCRL
Evaluation results
| Method | Links | |
|---|---|---|
| CRLTarget Model=Llama-3.1-8B-Instruct2026.02 | 11.7 | |
| CRLTarget Model=Gemma3-12B-IT2026.02 | 11.4 | |
| QCITarget Model=Llama-3.1-8B-Instruct2026.02 | 10.9 | |
| CRLTarget Model=GPT-4.1-Mini2026.02 | 10.9 | |
| QCITarget Model=Qwen3-30B-A3B-Instruct-25072026.02 | 10.3 | |
| CRLTarget Model=Qwen3-30B-A3B-Instruct-25072026.02 | 10.2 | |
| QCITarget Model=GPT-4.1-Mini2026.02 | 10.1 | |
| QCITarget Model=Gemma3-12B-IT2026.02 | 9.88 | |
| RSTarget Model=Llama-3.1-8B-Instruct2026.02 | 2.87 | |
| RSTarget Model=Gemma3-12B-IT2026.02 | 1.88 | |
| RSTarget Model=Qwen3-30B-A3B-Instruct-25072026.02 | 1.52 | |
| RSTarget Model=GPT-4.1-Mini2026.02 | 1.46 |