# stepfun-ai/gelab-zero

STEP-GUI: The top GUI agent solution in the galaxy.  Developed by the StepFun-GELab team and powered by StepFun’s cutting-edge research capabilities.

Repository: https://github.com/stepfun-ai/gelab-zero
Canonical: https://ross.abutalabs.com/products/gelab-zero
Homepage: https://opengelab.github.io/
Language: Python
License: MIT
License Family: permissive
Topics: agent, gui-agents, pua, phone-use-agent
Last push: 2026-05-11T05:50:07+00:00

## Health v2 (maintenance only)
Score: 52/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 81, release rhythm 35, longevity 19
- inputs: {"age_days": 278, "days_push": 114, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 2257, forks 198 (observed 2026-08-28T04:06:31.470411+00:00)

## What it is
GELab-Zero is an open-source GUI agent framework for mobile devices, combining a 4B vision-language model (GELab-Zero-4B) with engineering infrastructure for multi-device ADB management and task distribution. It runs fully locally with no cloud dependencies and includes an MCP server for integration with chat interfaces.

## Use cases
- automate android apps with a gui agent
- control phones via ai without app apis
- run a local phone-use agent with privacy
- manage adb tasks across multiple android devices
- benchmark gui agents on android world
- integrate a gui agent into chat via mcp

## When to choose
- you need a fully open-source, locally deployable mobile GUI agent with model and infrastructure included
- you want privacy-sensitive phone automation without cloud APIs
- you need multi-device Android agent orchestration or MCP integration

## When to avoid
- you need a mature commercial-grade agent with top benchmark scores
- you only target desktop or web GUIs rather than mobile
- you lack GPU resources to run the 4B vision-language model

## Facets
- artifact type: framework
- maturity: active
- function: agent-framework, machine-learning, computer-vision, mcp
- domain: mobile-development, artificial-intelligence
- platform: python, self-hosted
- tags: gui-agent, phone-use-agent, adb, mobile-automation, vision-language-model, android-world-benchmark, local-deployment, ai-agents, automation, android, docker

## Member repositories
- stepfun-ai/gelab-zero (main) score 52

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:06:31.470411+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T02:43:40.821743+00:00, confidence not recorded.
  - readme: https://github.com/stepfun-ai/gelab-zero (fetched 2026-08-28T04:06:31.470411+00:00, sha 585f5eb5b373)
  - homepage: https://opengelab.github.io/ (fetched 2026-08-29T10:23:32.440468+00:00, sha 3db2c669a16a)
- Data as of 2026-08-30T08:39:29.467469+00:00.