List your company

UI Action Grounding on ScreenSpot Desktop

74.2Success Rate (Text)CogAgent
2.9621.45539.9558.445May 8, 2026

Evaluation results

MethodLinks
CogAgentModel Size=18B2026.0574.2——20
SeeClickModel Size=9.6B2026.0572.2——30
Qwen2.5-VL-7B-Inst.Model Size=7B2026.0557.7——42.1
Skill-CMIB2026.0554.1——46.4
FuyuModel Size=8B2026.0533——3.6
GPT-4V2026.0520.2——11.8
MiniGPT-v2Model Size=7B2026.056.2——2.9
Qwen-VLModel Size=9.6B2026.055.7——5
FuyuSize=8B, Evaluation Protocol=Zero-shot2025.02—333.6—
GPT-4VSize=n/a, Evaluation Protocol=Zero-shot2025.02—20.211.8—
GPT-4V-OmniParserSize=n/a, Evaluation Protocol=Zero-shot, Detection Tool=OmniParser2025.02—92.749.4—
Magma-8BSize=8.6B, Evaluation Protocol=Zero-shot2025.02—75.352.9—
Qwen-VLSize=9.6B, Evaluation Protocol=Zero-shot2025.02—5.75—
SeeClickSize=9.6B, Evaluation Protocol=Zero-shot2025.02—72.230—