Project report / GitHub evidence

EvoCUA Model Card

这是可索引项目报告证据页:它保留 EvoCUA Model Card 的源材料入口、机制线索和限制提醒;正文仍需 reader/editor 与 academic public-copy review 后才能当作最终结论引用。

EvoCUA Model Card

FieldValue
Repositorymeituan/EvoCUA
CategoryComputer Use Agent 进化模型
Stars / forks snapshot317 / 23
LanguagePython
LicenseApache-2.0
Raw captureraw-github/meituan_evocua.md
Updated byhourly public metadata update, 2026-05-24

1. Role in Self Evolve

美团 EvoCUA 计算机使用 Agent 项目,发布 EvoCUA-32B/8B 并在 OSWorld、WindowsAgentArena 等 GUI 自动化评测上报告结果。

2. Working Principle

通过多模态 GUI 操作数据和评测反馈增强 computer-use agent 的跨环境执行能力。

3. Evidence Path

web GitHub page observed EvoCUA README, OSWorld/WAA claims, model release dates, and evaluation files; shell GitHub API was blocked by DNS and gh auth was invalid, so this run marks freshness as web-page observed rather than API verified.

4. Teaching Use

Use this card to explain whether a repository is a runtime, benchmark, harness-evolution loop, memory/skill substrate, or resource index. The key reading path is: raw capture -> classification row -> public site card -> project report -> aggregate GitHub analysis.

5. Limits

当前未克隆源码,未运行 benchmark;star/fork/commit 快照来自公开 GitHub 页面文本。