List your company

Mobile GUI Automation on AndroidWorld

76.9Overall Success RateAutoRPA
-3.07617.68738.4559.213Aug 15, 2025Sep 30, 2025Nov 15, 2025Jan 1, 2026Feb 16, 2026Apr 3, 2026May 20, 2026

Evaluation results

MethodLinks
AutoRPAModel=Claude-sonnet-4.5, Observation Type=Screenshot+A11y tree2026.0576.9—————36.1
ReAct†Model=Claude-sonnet-4.5, Observation Type=Screenshot+A11y tree2026.0576.7—————146
AutoRPAModel=GPT-52026.0575.9————4.3530.6
AutoRPAModel=GPT-5, Observation Type=Screenshot+A11y tree2026.0575.9—————30.6
ReAct†Model=GPT-52026.0574.1————8.57142.5
ReAct†Model=GPT-5, Observation Type=Screenshot+A11y tree2026.0574.1—————142.5
Mobile-Agent-v3Model=GUI-Owl-32B, Observation Type=Screenshot2026.0573.3——————
AutoRPA (code only)Model=GPT-52026.0570.7————2.726.2
AutoRPA (code only)Model=GPT-5, Observation Type=Screenshot+A11y tree2026.0570.7—————6.2
AutoRPAModel=Claude-sonnet-4.5, Observation Type=Screenshot2026.0570.1—————33.5
Gemini-2.5-CUModel=Gemini 2.5 Computer Use, Observation Type=Screenshot2026.0569.7——————
ReAct†Model=Claude-sonnet-4.5, Observation Type=Screenshot2026.0569—————145.1
Agent S3Model=GPT-5, Observation Type=Screenshot2026.0568.1——————
M3AModel=GPT-52026.0557————5.56164.6
M3AModel=GPT-5, Observation Type=Screenshot+A11y tree2026.0557—————164.6
CRAFT-GUI-32BTraining Stage=stage32025.0851.7——————
AutoRPAModel=GPT-4.12026.0551.7————1.8112.8
ReAct†Model=GPT-4.12026.0550————3.9168.7
M3AModel=GPT-4.12026.0548.3————2.23103.4
AutoRPA (code only)Model=GPT-4.12026.0547.2————1.422.7
UI-TARS-72B2025.0846.6——————
Zero-ShotBase Model=GPT-4o [20]2026.0345.2242.8666.6777.2729.09——
CRAFT-GUI-32BTraining Stage=stage22025.0844.8——————
CRAFT-GUI-32BTraining Stage=stage12025.0841.4——————
SeeActModel=GPT-52026.0540.3————12.43134.5
SeeActModel=GPT-5, Observation Type=Screenshot+A11y tree2026.0540.3—————134.5
UI-Copilot-7Btool calling=true2026.0439.1——————
AutoRPAModel=GPT-4o2026.0537————1.7614.7
ReAct†Model=GPT-4o2026.0535.2————4.9779.9
Qwen2.5-VL-72B2025.0835——————
GPT-4o2025.0834.5——————
GPT-4o (SoM)2026.0434.5——————
M3AModel=GPT-4o2026.0534.5————2.83120.2
AutoRPA (code only)Model=GPT-4o2026.0534.3————1.082.6
UI-S1-7B2026.0434——————
UI-TARS-7B2026.0433——————
UI-Copilot-7Btool calling=false2026.0432.2——————
ClaudeVersion=Claude-computer-use2025.0827.9——————
Claude-CU (SoM)2026.0427.9——————
Aguvis-72B2025.0826.1——————
SeeActModel=GPT-4.12026.0525.4————5.1458.8
HATSBase Model=InternVL2-8B [4]2026.0324.3516.0925.9353.0313.58——
HATSBase Model=Qwen2-VL-7B [61]2026.0324.3518.3930.7751.529.88——
HATSBase Model=InternVL2-4B [4]2026.0322.617.2438.4640.918.64——
Qwen2.5-VL-32B2025.0822——————
Qwen2.5VL-7B2026.0422——————
OS-Genesis-7B2026.0417.4——————
AgentCPM-GUI-8B2026.0416.4——————
UI-Copilot-3B2026.0415.7——————
SeeActModel=GPT-4o2026.0515.3————6.3868.4
UI-R1-7B2026.0415.1——————
OS-GenesisBase Model=InternVL2-8B [4]2026.0314.8113.79047.624.08——
OS-Atlas-7B2026.0412.1——————
OS-GenesisBase Model=Qwen2-VL-7B [61]2026.0311.517.86031.821.85——
OS-GenesisBase Model=InternVL2-4B [4]2026.0311.37.1411.1136.363.64——
Task-DrivenBase Model=Qwen2-VL-7B [61]2026.0310.438.333.730.32.47——
UI-R1-3B2026.048.2——————
Task-DrivenBase Model=InternVL2-4B [4]2026.037.93.49019.73.7——
ShowUI-2B2026.047——————
Task-Driven w. Self-InstructBase Model=InternVL2-8B [4]2026.036.255.95013.643.09——
Task-Driven w. Self-InstructBase Model=InternVL2-4B [4]2026.036.092.33.721.210.62——
Task-Driven w. Self-InstructBase Model=Qwen2-VL-7B [61]2026.036.096.97.4125.760.62——
Qwen2.5VL-3B2026.045——————
Task-DrivenBase Model=InternVL2-8B [4]2026.033.481.15012.120.62——
Zero-ShotBase Model=Qwen2-VL-7B [61]2026.031.75007.580——
Zero-ShotBase Model=InternVL2-8B [4]2026.031.04003.130——
Zero-ShotBase Model=InternVL2-4B [4]2026.031.02002.380——
Qwen2VL-2B2026.040——————