List your company

Multimodal Agent Performance on Online Deployment Suite Aggregated

75.3AccuracyToolGate
70.651271.858173.06574.2719Jun 2, 2026

Evaluation results

MethodLinks
ToolGateBackbone=235B-FP8, Training regime=Cross-domain2026.0675.3690.2
BaselineBackbone=235B-FP8, Training regime=None2026.0675.221001.34
ToolGateBackbone=30B, Training regime=In-domain2026.0672.48671.02
ToolGateBackbone=30B, Training regime=Cross-domain2026.0671.15641
BaselineBackbone=30B, Training regime=None2026.0670.831002.73