List your company

GUI Agent Task on MemGUI-Bench Medium

35.7Precision@1M3A
-1.4288.21117.8527.489Jun 18, 2026

Evaluation results

MethodLinks
M3AMemory Type=Memory Agent2026.0635.75048.7
MemGUI-Agent-235BMemory Type=ConAct, Protocol=zero-shot, Backbone=Qwen3-VL-235B-Thinking2026.0635.761.952.3
Qwen3-VL-235B-ThinkingMemory Type=Action-Thought2026.0628.647.638.8
MemGUI-8B-SFTMemory Type=ConAct, Protocol=fine-tuned, Backbone=Qwen3-VL-8B-Instruct2026.0623.833.336.5
Qwen3-VL-235B-InstructMemory Type=Action-Thought2026.0621.440.528.8
Agent-S2Memory Type=Memory Agent2026.061942.942.5
T3AMemory Type=Memory Agent2026.0616.745.2—
Qwen3-VL-32B-InstructMemory Type=Action-Thought2026.0614.323.822.4
GUI-Owl-1.5-32B-InstructMemory Type=Action-Thought2026.067.114.321.3
Qwen3-VL-8B-InstructMemory Type=Action-Thought2026.064.814.316
Mobile-Agent-EMemory Type=Memory Agent2026.062.42.44.4
GUI-Owl-1.5-8B-InstructMemory Type=Action-Thought2026.062.42.416.5
Mobile-Agent-V2Memory Type=Memory Agent2026.06000
SeeActMemory Type=Rule-based2026.0602.40.5
AppAgentMemory Type=Action-Thought2026.0602.40
GUI-Owl-7BMemory Type=Action-Thought2026.0602.45
UI-TARS-1.5-7BMemory Type=Multi-turn2026.06001.7
UI-Venus-7BMemory Type=Action-Thought2026.06001.2
CogAgentMemory Type=No History2026.06000