General Assistant Task Solving on GAIA (unseen)
13.21Level 1 ScoreOFA-MAS
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| OFA-MASBase Model=GPT-4o-mini2026.01 | 13.21 | 9.3 | 3.85 | 8.79 | |
| ChainBase Model=GPT-4o-mini, Topology=Static (Chain)2026.01 | 11.32 | 6.98 | 3.85 | 7.38 | |
| AgentPruneBase Model=GPT-4o-mini2026.01 | 9.43 | 8.14 | 0 | 5.86 | |
| EIB-LEARNERBase Model=GPT-4o-mini2026.01 | 9.43 | 6.98 | 0 | 5.47 | |
| CompleteBase Model=GPT-4o-mini, Topology=Static (Complete)2026.01 | 5.66 | 8.14 | 0 | 4.6 | |
| G-DesignerBase Model=GPT-4o-mini2026.01 | 5.66 | 6.98 | 3.85 | 5.5 |