List your company

Trajectory completion judgment on OGRBench Ubuntu

88.7AccuracyStainFlow
55.31663.98372.6581.317Jun 5, 2026

Evaluation results

MethodLinks
StainFlowVerifier=Gemini-3-Flash2026.0688.789.4
StainFlowVerifier=Qwen3.5-27B2026.0687.288.1
OS-ThemisVerifier=Qwen3.5-27B2026.0686.887.2
ZeroGUIVerifier=Qwen3.5-27B2026.0686.587.5
OS-ThemisVerifier=Gemini-3-Flash2026.0686.587
StainFlowVerifier=Qwen3.5-9B2026.0685.386.4
StainFlowVerifier=GPT-52026.0684.185.8
ZeroGUIVerifier=GPT-52026.0681.983.2
StainFlowVerifier=Qwen3-VL-8B2026.0681.984.5
ZeroGUIVerifier=Gemini-3-Flash2026.0681.882.6
ZeroGUIVerifier=Qwen3.5-9B2026.0681.183.7
OS-ThemisVerifier=GPT-52026.0679.878.4
OS-ThemisVerifier=Qwen3.5-9B2026.0679.479.1
ZeroGUIVerifier=Qwen3-VL-8B2026.0679.182.7
OS-ThemisVerifier=Qwen3-VL-8B2026.0677.376.2
DigiRLVerifier=Gemini-3-Flash2026.0663.954.2
DigiRLVerifier=Qwen3.5-27B2026.0662.451.5
DigiRLVerifier=GPT-52026.0662.453
DigiRLVerifier=Qwen3-VL-8B2026.0660.250.3
DigiRLVerifier=Qwen3.5-9B2026.0656.637.6