cd /news/artificial-intelligence/mintact-a-unified-visual-agent-for-d… · home topics artificial-intelligence article
[ARTICLE · art-135983] src=arxiv.org ↗ pub= topic=artificial-intelligence verified=true sentiment=↑ positive

MintAct: A Unified Visual Agent for Digital Environments

Researchers submitted MintAct, a family of vision-language models trained at 2B, 4B, and 8B scales that unifies UI grounding, multi-step navigation across mobile, desktop, and web, and visual tool use, to arXiv on 18 Sep 2026. MintAct matches per-domain specialists across these capabilities and achieves state-of-the-art performance of 48.9 on OSWorld-Verified, supported by a scalable environment and asynchronous reinforcement learning infrastructure hosting hundreds of concurrent instances across heterogeneous per-domain backends.

read2 min views3 publishedSep 21, 2026
MintAct: A Unified Visual Agent for Digital Environments
Image: source
  [Submitted on 18 Sep 2026]


[View PDF](http://arxiv.org/pdf/2609.22083v1)

[HTML (experimental)](https://arxiv.org/html/2609.22083v1)

Abstract:We present MintAct, a family of vision-language models that unifies UI grounding, multi-step navigation across mobile, desktop, and web, and visual tool use, trained at 2B, 4B, and 8B scales. Through careful design of our environments, data, and training recipes, MintAct models match the performance of per-domain specialists across all of these capabilities. To enable this, we develop a scalable environment and reinforcement learning (RL) infrastructure. On the environment side, we host hundreds of concurrent instances across heterogeneous per-domain backends, serving both trajectory data collection and online RL. To enable efficient and scalable RL training, an asynchronous framework keeps explicit control over the cross-domain training distribution and remains stable under noisy environment feedback and off-policy drift. Experimental results show that MintAct achieves state-of-the-art performance (48.9 on OSWorld-Verified) across a wide range of benchmarks at comparable model sizes.

References & Citations

...

Bibliographic Explorer

(What is the Explorer?) Connected Papers

(What is Connected Papers?) Litmaps

(What is Litmaps?) scite Smart Citations

(What are Smart Citations?) alphaXiv

(What is alphaXiv?) CatalyzeX Code Finder for Papers

(What is CatalyzeX?) DagsHub

(What is DagsHub?) Gotit.pub

(What is GotitPub?) Hugging Face

(What is Huggingface?) ScienceCast

(What is ScienceCast?) Influence Flower

(What are Influence Flowers?) CORE Recommender

(What is CORE?) arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.

Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.

Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.

── more in #artificial-intelligence 4 stories · sorted by recency
── more on @mintact 3 stories trending now
sponsored brought to you by zahid.host 4,200+ EU-deployed projects
reading about agents? ship yours in a single git push.

Run your AI side-project on zahid.host

EU-based hosting, git-push deploys, automatic HTTPS, no cold starts. Free tier with a custom domain — perfect for shipping the agent you just read about.

$git push zahid main
Live at https://your-agent.zahid.host
Get free account → Pricing
from €0/mo · no card required
LIVE [news/mintact-a-unified-vi…] indexed:0 read:2min 2026-09-21 ·