List your company

Multimodal Agent Performance on HR-Bench 8k

78.2AccuracyBaseline
72.37673.88875.476.912Jun 2, 2026

Evaluation results

MethodLinks
BaselineBackbone=235B-FP8, Training regime=None2026.0678.2
ToolGateBackbone=235B-FP8, Training regime=Cross-domain2026.0677
ToolGateBackbone=30B, Training regime=In-domain2026.0675.5
ToolGateBackbone=30B, Training regime=Cross-domain2026.0674.3
BaselineBackbone=30B, Training regime=None2026.0672.6