List your company

GUI Agent Task on MemGUI-Bench (Overall)

37.5Precision@1MemGUI-Agent-235B
-1.58.62518.7528.875Jun 18, 2026

Evaluation results

MethodLinks
MemGUI-Agent-235BMemory Type=ConAct, Protocol=zero-shot, Backbone=Qwen3-VL-235B-Thinking2026.0637.562.546.8
M3AMemory Type=Memory Agent2026.0632.847.739.3
Agent-S2Memory Type=Memory Agent2026.0627.349.239.5
Qwen3-VL-235B-ThinkingMemory Type=Action-Thought2026.0624.246.930
Qwen3-VL-235B-InstructMemory Type=Action-Thought2026.0623.436.729.2
MemGUI-8B-SFTMemory Type=ConAct, Protocol=fine-tuned, Backbone=Qwen3-VL-8B-Instruct2026.0623.435.930.2
T3AMemory Type=Memory Agent2026.0622.742.229.6
Qwen3-VL-32B-InstructMemory Type=Action-Thought2026.0620.33225
GUI-Owl-1.5-8B-InstructMemory Type=Action-Thought2026.0611.715.615.5
GUI-Owl-1.5-32B-InstructMemory Type=Action-Thought2026.0610.920.318.4
Qwen3-VL-8B-InstructMemory Type=Action-Thought2026.069.420.315.1
GUI-Owl-7BMemory Type=Action-Thought2026.066.210.25.7
Mobile-Agent-EMemory Type=Memory Agent2026.065.510.22.4
UI-Venus-7BMemory Type=Action-Thought2026.065.57.82.6
Mobile-Agent-V2Memory Type=Memory Agent2026.063.13.90
AppAgentMemory Type=Action-Thought2026.063.19.41.5
UI-TARS-1.5-7BMemory Type=Multi-turn2026.063.16.23.8
SeeActMemory Type=Rule-based2026.062.35.50.2
CogAgentMemory Type=No History2026.06000