List your company

OS GUI Agentic Task Execution on OSWorld 361 tasks (Verified)

79.17OS Success RateGTA1 w/ GPT-5
2.033222.059142.08562.1109Jan 12, 2026Jan 22, 2026Feb 2, 2026Feb 13, 2026Feb 23, 2026Mar 6, 2026Mar 17, 2026

Evaluation results

MethodLinks
GTA1 w/ GPT-5Step=1002026.0179.1763.9162.5679.5950.9163.41
OS-SYMPHONY w/ GPT-5Step=1002026.0179.1765.7367.7669.2357.9865.84
Agent S3 w/ GPT-5Step=1002026.0177.566.4661.2369.851.3762.63
CoAct-1 w/ GPT-5Step=1002026.017562.9357.9471.4347.8759.93
CoAct-1 w/ GPT-5Step=1502026.017562.9361.7871.4347.8760.76
OS-SYMPHONY w/ GPT-5Step=502026.017564.8561.1969.2354.8663.61
UiPath w/ GPT-5Step=502026.0173.9149.5262.1271.4337.353.69
OS-SYMPHONY w/ GPT5-MiniStep=502026.0173.6858.1761.397547.3758.05
Claude-Sonnet-4.5Step=1002026.0170.8372.5961.3563.2749.5462.84
CoAct-1 w/ GPT-5Step=502026.0170.8360.6554.0969.3942.3756.39
OS-SYMPHONY w/ Qwen3-VL-32B-ThinkingStep=502026.0170.8340.9766.0473.0831.2250.23
DeepMiner-Mano-72BStep=1002026.0166.6763.2252.5183.6724.4153.91
Claude 4.5 SonnetParams=-2026.0362.9—————
Agent S3 w/ GPT-5-MiniStep=50, reproduced_by_paper=true2026.0162.554.6246.6744.937.0447.58
OpenCUA-72BStep=1002026.0161.1344.7349.9572.5822.1644.91
OS-SYMPHONY w/ Qwen3-VL-32B-InstructStep=502026.0158.3340.9453.5475.131.2446.86
Agent S2.5 w/ o3Params=7B w/ -2026.0356—————
GTA1-32B w/ o3Params=32B w/-2026.0355.4—————
UI-TARS-2Params=-2026.0353.1—————
GTA1-7B w/ o3Params=7B w/-2026.0353.1—————
Jedi-7B w/ o3Params=7B w/ -2026.0350.2—————
DeepMiner-Mano-7BStep=1002026.015039.2844.8773.4717.240.15
Agent S3 w/ Qwen3-VL-32B-InstructStep=50, reproduced_by_paper=true2026.015036.6750.6261.2221.9640.11
Agent S2 w/ GPT-5Params=7B w/ -2026.0348.8—————
Qwen3-VL-32B-Thinking w/ GPT-4.1Params=32B w/ -2026.0343.2—————
UI-TARS-1.5Params=-2026.0341.8—————
UI-TARSStep=1002026.0141.6750.4255.6951.0214.6641.85
UI-TARS-2Step=1002026.0141.6761.1162.1261.2234.1353.1
Claude 4 SonnetParams=-2026.0341.4—————
Qwen3-VL-32B-Thinking w/ TraceR1Params=32B w/ 8B2026.0341.2—————
OpenAI CUA-o3Params=-2026.0338.1—————
Qwen3-VL-235B-A22B-ThinkingParams=236B2026.0338.1—————
Seed1.5-VLParams=-2026.0336.7—————
Qwen3-VL-32B-ThinkingParams=32B2026.0335.6—————
OpenCUA-32BParams=32B2026.0334.8—————
Qwen3-VL-8B-ThinkingParams=8B2026.0333.9—————
UI-TARS-1.5-7B w/ GPT-4.1Params=7B w/ -2026.0331.6—————
UI-TARS-1.5-7B w/ TraceR1Params=7B w/ 8B2026.0330.9—————
UI-TARS-1.5-7BParams=7B2026.0327.4—————
UI-TARS-72B-DPOParams=72B2026.0327.1—————
OpenCUA-7BParams=7B2026.0326.6—————
o3Params=-2026.0323—————
Qwen2.5-vl-72bParams=72B2026.035—————
Qwen3-VL-32B-InstructStep=502026.01—————32.4
Qwen3-VL-32B-ThinkingStep=502026.01—————41