xlang-ai / CUA-Gym
Scalable pipeline for synthesizing verifiable RLVR training data for computer-use agents
Why this job: works in the areas we cover, but does not say which job it does
What it does
CUA-Gym is a scalable pipeline for synthesizing verifiable RLVR training data for computer-use agents (CUAs). Given a topic, it jointly produces task instructions, environment states, and reward functions as verified triples — using coding agents to handle the engineering work previously requiring human experts.
Get started
No single install command. Read the steps in the README (opens in new tab)
- Maintenance
- Active
- Last commit 5 months ago
- Stars
- 228
- +2 this week
- Licence
- Custom
- Read before business use
- Downloads for
- None listed
- No release files
- Latest version
- None yet
- No releases
- Built with
- Python
- 100% of the code
Is it looked after?
- Contributors
- Not stated
- Last 90 days
- Issues
- 11% closed
- 8 still open
- Pull requests
- 0 merged
- 1 waiting
- Releases
- None
- Never published
- Contributing guide (opens in new tab)
- No security policy
- No code of conduct
Similar repos
| Repo | Stars | Last commit | Licence | Downloads for |
|---|---|---|---|---|
| Tencent-Hunyuan / ArtifactsBenchmark | 282 | 10 months ago | Custom | No downloads |
| CraftOS-dev / CraftBot | 422 | 8 days ago | MIT | Mac, Windows, Linux |
More: languages, topics and project record
Languages
- Python 100%
Topics
- agent
- computer-use-agent
- data-synthesis
- llm
- osworld
- python
- qwen
- reinforcement-learning
- reward-model
- rlvr
Project record
- JobOther AI tools
- StatusListed
- Commits31
- Forks22
- Watching2
- Stars in 28 days+13
- CreatedJan 6, 2026
- First seen by usNov 7, 2026
- CheckedNov 8, 2026