---
format: "aidr-story-markdown/v1"
id: "85a81bd791d18b545ef3372f0bb3c69ee838fa366e61049043875688b2cef8b9"
canonical_url: "https://aidr.today/85a81bd7?lang=en"
title: "Cua Debuts First RLOO Decision Model for Computer Use"
lang: "en"
requested_lang: "en"
available_langs: ["en","vi"]
translation_fallback: null
fallback_fields: []
published_at: "2026-09-23T23:36:35.000Z"
category: "Models"
topics: ["cua","rloo","decision-model","computer-use","fine-tuning","agent","inference","automation"]
source_urls: ["https://huggingnews.com/ai/cua-debuts-first-rloo-decision-model-for-computer-use-e106a75a","https://marketbrief.now/ai/cua-debuts-first-rloo-decision-model-for-computer-use-e106a75a"]
summary: "Cua released the Cua-S1-4B-0.2, a 4B parameter multimodal decision model trained with RLOO for automating computer tasks, with a browser-based demo available on Hugging Face."
---

# Cua Debuts First RLOO Decision Model for Computer Use

> [Open the canonical story](<https://aidr.today/85a81bd7?lang=en>)

**Published:** 2026-09-23T23:36:35.000Z
**Category:** Models
**Topics:** cua, rloo, decision\-model, computer\-use, fine\-tuning, agent, inference, automation

## Summary

Cua released the Cua\-S1\-4B\-0\.2, a 4B parameter multimodal decision model trained with RLOO for automating computer tasks, with a browser\-based demo available on Hugging Face\.

## Sources

- [Story source](<https://huggingnews.com/ai/cua-debuts-first-rloo-decision-model-for-computer-use-e106a75a>)
- [Story source](<https://marketbrief.now/ai/cua-debuts-first-rloo-decision-model-for-computer-use-e106a75a>)

