Dynamic Interactive Conversation on GOODAILTM (Isolated)
9.2Solved ScenariosOBLIVION
Evaluation Results
| Method | Links | |
|---|---|---|
| OBLIVIONModel=GPT-4.1-mini2026.03 | 9.2 | |
| OBLIVIONModel=GPT-4o-mini2026.03 | 9 | |
| BEYONDPROMPTSModel=GPT-4o-mini, Type=memory2026.03 | 8.87 | |
| BEYONDPROMPTSModel=GPT-4.1-mini, Type=memory2026.03 | 8.53 | |
| FULLCTXModel=GPT-4.1-mini, Type=direct2026.03 | 8.11 | |
| OBLIVIONModel=Qwen3-30B-A3B†2026.03 | 7.91 | |
| FULLCTXModel=GPT-4o-mini, Type=direct2026.03 | 7.22 | |
| FULLCTXModel=Qwen3-30B-A3B†, Type=direct2026.03 | 6.49 | |
| OBLIVIONModel=Phi-4-mini†2026.03 | 6.45 | |
| BEYONDPROMPTSModel=Qwen3-30B-A3B†, Type=memory2026.03 | 6.45 | |
| BEYONDPROMPTSModel=Phi-4-mini†, Type=memory2026.03 | 5.57 | |
| FULLCTXModel=Phi-4-mini†, Type=direct2026.03 | 4.44 |