Congrats to @trycua on open-sourcing cua-bench!

We're collaborating on task design & data curation for GUI workflows – bringing the same systematic evaluation approach from Terminal-Bench to computer-use agents.

15 native tasks / 40 variations + OSWorld & Windows Agent Arena.… https://t.co/7GZkGHHTYx

— Snorkel AI (@SnorkelAI) 23 janvier 2026

Congrats to @trycua on open-sourcing cua-bench! We're collaborating on task design & data curation for GUI workflows – bringing the same systematic evaluation approach from Terminal-Bench to computer-use agents. 15 native tasks / 40 variations + OSWorld & Windows Agent Arena.

→ View original post on X — @snorkelai,