# Is CUA-Master-27B SOTA?

Pending benchmarks. Announced Nov 2, 2026; no independent results yet.

As of Nov 22, 2026, CUA-Master-27B is pending benchmarks on isitsota: announced Nov 2, 2026, with no independent results on the boards we track yet.

A 27B computer-use agent model from CUHK MMLab and collaborators, built on Qwen3.5-27B for GUI-only and hybrid GUI + CLI tasks and trained with iterative SFT and RL on verified trajectories mined from instructional videos and synthesized tasks. Weights on Hugging Face; no license stated.

Canonical: https://isitsota.com/m/cua-master-27b
Last updated: 2026-11-03
Tracking since: 2026-11-03

## Announced facts

| Fact | Value |
| --- | --- |
| Lab | CUHK MMLab |
| Announced | Nov 2, 2026 |
| API | Not yet · None announced |
| Open weights | Published · Hugging Face yxl66666/CUAMaster-27B (an author's account, linked from the paper), BF16 safetensors, not gated |
| Parameters | 27B (27,356,728,560 stored BF16 parameters) |
| Type | Computer-use agent model for GUI-only and hybrid GUI + CLI workflows |
| Base model | Qwen3.5-27B |
| Training | Iterative SFT and RL (a GRPO-style online RL setup following ScaleCUA) on verified trajectories from tasks mined from instructional videos and from compositional task synthesis |
| License | Not stated on the model card or the code repository |
| Code | github.com/xlyu0106/CUA-Master: data pipeline (video harvesting, trajectory reconstruction, task synthesis, sandboxed trajectory collection) and SFT data preparation |
| Paper | CUA-Master: Scaling Computer-use Agents through Open-World Recursive Improvements (arXiv 2611.03126) |
| Authors | Xinlei Yu and 17 co-authors from CUHK MMLab, Tsinghua University, HKUST (Guangzhou) and JD Explore Academy |
| Requirements | A computer-use execution environment with GUI and/or CLI actions; the card advises a sandbox |

Tags marked unsure are not stated on the lab's own pages.

## What would let us call it

Verdicts use only independent boards we already track. Two consecutive snapshots are required before any crown.

- [x] Publicly announced with a source
- [x] Usable by the public (API or weights)
- [ ] Listed on an independent board we track
- [ ] Enough results for a rank in a category
- [ ] Holds across two snapshots (crowns only)

## What the lab claims

Self-reported. Shown for context; not used in any verdict.

- Lab claim: OSWorld Verified 80.2 and OSWorld 2.0 36.9 (GUI), WeaveBench 43.9 and Agents' Last Exam 25.0 full-pass rate (GUI + CLI).
- Lab claim: WindowsAgentArena 56.2, MacOSWorld 43.0 and MacAgentBench 59.9.
- Lab claim: average time per task of 8 min on OSWorld Verified, 1 h 7 min on OSWorld 2.0, 34 min on WeaveBench and 52 min on Agents' Last Exam.

## Sources

- PAPER: [CUA-Master: Scaling Computer-use Agents through Open-World Recursive Improvements (arXiv)](https://arxiv.org/abs/2611.03126) (2026-11-02)
- LAB: [First author Xinlei Yu on X](https://x.com/vajkcgvd/status/2117559546852114763) (2026-11-03)
- LAB: [CUA-Master project page](https://xlyu0106.github.io/CUA-Master/)
- REPO: [xlyu0106/CUA-Master on GitHub](https://github.com/xlyu0106/CUA-Master)
- REPO: [yxl66666/CUAMaster-27B on Hugging Face](https://huggingface.co/yxl66666/CUAMaster-27B)

Tracking continues until an independent board lists it.

## Related

- Lab: [CUHK MMLab](https://isitsota.com/labs/cuhk-mmlab.md)
- All pending releases: https://isitsota.com/releases.md
- JSON: https://api.isitsota.com/v1/models/cua-master-27b
