Share your thoughts, 1 month free Claude Pro on usSee more

Home/Benchmarks

Computer-use on GAIA L1

40.86Headline SR

OS-Copilot/FRIDAY

Updated 8d ago

Evaluation Results

Method	Links
OS-Copilot/FRIDAY 2026.05		40.86

SOTA Paper

OS-Copilot/FRIDAY

OS-Copilot: Towards Generalist Computer Agents with Self-Improvement

Dataset

GAIA

Follow for update

@wizwand_team Discord

Related Benchmarks

Web Navigation and Task Automation on VisualWebArena 910 tasks (Full)Visual Web Navigation on VWA-910 Web Navigation on VisualWebArena Computer Use on OSWorld (Headline SR)Computer-use on OSWorld 50-step

© 2026 wizwand

Blog Contact Changelog Swarm

Privacy Policy Terms of Service FAQs Swarm Docs