UI-Venus-2 (arXiv 2609.00028, Wed 2 Sep digest, 31 authors under the Venus Team banner) is a general-purpose, open-source foundation GUI agent spanning mobile, web, and desktop through a unified closed-loop reasoning-action framework. It scales three axes jointly: environments (170+ multilingual mobile apps plus native desktop operating systems), tasks (a deep-research pipeline generating function-grounded instructions), and verification (trace-level and sample-level evaluators combining visual keypoints and multi-model voting, to make RL training signals reliable). It adds safety-aware mechanisms around consequential actions and is positioned as an open foundation for more generalizable, verifiable, self-reflective agents.
UI-Venus-2: an open-source foundation GUI agent for mobile, web, and desktop
2026-09-03Frontier watchTRIAL
UI-Venus-2 is an open GUI-agent foundation spanning mobile, web and desktop with one closed-loop reasoning-action framework. Its repository, 9B weights, training and evaluation tools are now public. The model card reports 170+ multilingual mobile apps and trace- and sample-level verification, but explicitly says the model-weight license is still pending confirmation.
Imported from the existing research archive. Migration preserves the recorded judgment and is not a fresh verification.
What this means for you
For self-hosted computer-use teams, the public weights and framework make a real evaluation possible. The unresolved model license blocks commercial adoption, so trial only in an isolated non-production environment.
FRONTIER WATCH
Public artifacts and validated results are different.
The recorded fact is the public work and its artifacts. Author-reported outcomes do not establish independent reproduction. Next checks include usable code, task conditions and third-party validation.
Evidence & related material
UI-Venus-2's 9B weights and framework are public, but the model card explicitly says the weight license remains pending.
Subsequent changes
The 9B weights and framework are public, but the license remains pending; the rating moves from WATCH to TRIAL.
Technical details & scores
| Arxiv | 2609.00028, announced in the Wed 2 Sep 2026 digest (submitted 8/27) |
|---|---|
| Scope | general-purpose foundation GUI agent: mobile + web + desktop via a unified closed-loop reasoning-action framework |
| Environments | 170+ multilingual mobile apps plus native desktop operating systems |
| Tasks | deep-research pipeline generating function-grounded instructions |
| Verification | trace-level and sample-level evaluators: visual keypoints + multi-model voting, aimed at reliable RL signals |
| Safety | safety-aware mechanisms around consequential actions |
| Openness | repository, 9B weights, training and evaluation tools are public |
| License | model-weight license pending final confirmation; repository has no declared license |
Technical impact 3
Engineering value 3
Adoption signal 3
Maturity 3
Verification cost 3
Risk 4