List your company

GUI Web Agent Navigation on Mind2Web Online

67.3Overall Average ScoreOpenWebRL-4B w/ OpenWebRL-Judge-8B
2.319.17536.0552.925Feb 25, 2026Mar 13, 2026Mar 29, 2026Apr 14, 2026Apr 30, 2026May 16, 2026Jun 1, 2026

Evaluation results

MethodLinks
OpenWebRL-4B w/ OpenWebRL-Judge-8Bbackbone=4B, # Steps=30, # Tasks=2.2k2026.0667.3————————
OpenWebRL-4Bbackbone=4B, # Steps=30, # Tasks=2.2k2026.0667————————
OpenWebRL-8Bbackbone=8B, # Steps=30, # Tasks=2.2k2026.0667————————
Qwen3-VL-235B-A22B-Thinking# Steps=30, # Tasks=–2026.0663.7————————
OpenAI computer-use-preview# Steps=100, # Tasks=–2026.0658.3————————
GPT-5 (SoM)# Steps=100, # Tasks=–2026.0657.7————————
Gemini computer-use-preview# Steps=100, # Tasks=–2026.0657.3————————
o3 (SoM)# Steps=100, # Tasks=–2026.0655.4————————
OpenWebRL-8B-SFTbackbone=8B, # Steps=30, # Tasks=0.9k2026.0654————————
OpenWebRL-4B-SFTbackbone=4B, # Steps=30, # Tasks=0.4k2026.0647————————
Gemini-3-flash (Axtree)# Steps=100, # Tasks=–2026.0644.8————————
GPT-5 (Axtree)# Steps=30, # Tasks=–2026.0641.9————————
Qwen3-VL-8B-Thinking# Steps=30, # Tasks=–2026.0638.7————————
MolmoWeb-8B# Steps=100, # Tasks=>278.5k2026.0635.3————————
Gemini-3-flash (Axtree)# Steps=30, # Tasks=–2026.0634.8————————
GPT-4o (SoM)# Steps=100, # Tasks=–2026.0634.6————————
Fara-7B# Steps=100, # Tasks=>123.2k2026.0634.1————————
GLM-4.1V-9B-Thinking# Steps=100, # Tasks=–2026.0633.9————————
Qwen3-VL-4B-Thinking# Steps=30, # Tasks=–2026.0632————————
UI-TARS-1.5-7B# Steps=100, # Tasks=–2026.0631.3————————
MolmoWeb-4B# Steps=100, # Tasks=>278.5k2026.0631.3————————
GPT-4.1 + UGround-v1-7BSteps=302026.0229.741.321.75.222.747.53528.636.7
GPT-5 + UGround-v1-7BSteps=302026.0229.74024.514.3264531.524.733.3
GUI-Libra-8BSteps=302026.022831.317.510.419.342.537.828.636.7
Qwen3-VL-32BSteps=302026.0226.833.817.57.819.34531.528.634.3
GUI-Libra-4BSteps=302026.0225.736.318.26.5204530.119.531.3
GUI-Libra-7BSteps=302026.0225.536.315.42.617.747.530.823.433.3
GPT-4o + UGround-v1-7BSteps=302026.0224.338.8146.518.74527.319.530
ScaleCUA-7BSteps=302026.0223.733.814.73.91747.527.318.230.3
ScaleCUA-32BSteps=302026.0223.531.314.76.51743.829.416.930
Qwen3-VL-4BSteps=302026.0221.733.810.56.515.743.823.818.227.7
GUI-Libra-3BSteps=302026.0221.328.89.85.213.747.52124.729
Qwen3-VL-8BSteps=302026.0219.323.89.801143.823.119.527.7
ScaleCUA-3BSteps=302026.0216.3304.92.61137.518.211.721.7
Qwen2.5-VL-7BSteps=302026.0215.822.57.709.736.318.91322
Qwen2.5-VL-32BSteps=302026.0213.512.57.71.37.328.814.719.519.7
Qwen2.5-VL-3BSteps=302026.024.83.801.31.316.37.71.38.3