Multimodal Agent Performance on Online Deployment Suite Aggregated
75.3AccuracyToolGate
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| ToolGateBackbone=235B-FP8, Training regime=Cross-domain2026.06 | 75.3 | 69 | 0.2 | |
| BaselineBackbone=235B-FP8, Training regime=None2026.06 | 75.22 | 100 | 1.34 | |
| ToolGateBackbone=30B, Training regime=In-domain2026.06 | 72.48 | 67 | 1.02 | |
| ToolGateBackbone=30B, Training regime=Cross-domain2026.06 | 71.15 | 64 | 1 | |
| BaselineBackbone=30B, Training regime=None2026.06 | 70.83 | 100 | 2.73 |