Skip to content
Back to search

LumaBrowser

Local AI desktop app: a Chromium browser with a GPU-aware llama.cpp runtime, OpenAI-compatible API, AI chat and agents, Stable Diffusion image/video, TTS and music, plus an MCP server so Claude Code, Codex and Cline can drive your browser. Windows, macOS, Linux.

Safety scan

A static text scan, not a guarantee. Select the badge to see what was scanned and every finding.

Checked yesterday

Check it deeper on your machine

It has changed since its scan on 18 Nov 2026.

As a Claude Code plugin, install it and then ask Claude Code:

/plugin marketplace add keithmackay/mackayi/plugin install scanskillsafety@mackayiDo a deep scan of https://github.com/amurgola/LumaBrowser with scanskillsafety

Or from a terminal, clone it and run it:

git clone https://github.com/keithmackay/scanskillsafety-skill ~/.claude/skills/scanskillsafetynode ~/.claude/skills/scanskillsafety/cli.cjs --deep 'https://github.com/amurgola/LumaBrowser'

It checks the exact commit: every file, its packages against OSV (api.osv.dev), how it compares with this page's scan (findsafeskills.com receives the repository's address), and, if you agree, Claude reading every file with your own Claude subscription. It never runs the code.

Badge for your README
findsafeskills safety rating

Markdown

[![findsafeskills safety rating](/api/listings/amurgola--lumabrowser/badge.svg)](/listings/amurgola--lumabrowser)

HTML

<a href="/listings/amurgola--lumabrowser"><img src="/api/listings/amurgola--lumabrowser/badge.svg" alt="findsafeskills safety rating" /></a>

Install

i

What was scanned

The files the safety scan read, as written by the repository's author and shown as plain text. Lines with a finding are marked.

README.md · README · 219 lines
  1. <p align="center">
  2. <img src="icon/icon.png" alt="" width="88" height="88">
  3. </p>
  4. <h1 align="center">LumaBrowser</h1>
  5. <p align="center">
  6. <b>Your local AI stack in one app.</b><br>
  7. A web browser that runs language, image, video and voice models on your own
  8. GPUs, and lets agents work in the tabs you are already signed in to.
  9. </p>
  10. <p align="center">
  11. <a href="https://github.com/amurgola/LumaBrowser/releases/latest"><b>Download</b></a> ·
  12. <a href="Documentation/features/README.md">Features</a> ·
  13. <a href="https://lumabyte.com/apis">API docs</a> ·
  14. <a href="https://lumabyte.com">Website</a>
  15. </p>
  16. <p align="center">
  17. Free · Open source (AGPL-3.0) · Windows, macOS and Linux
  18. </p>
  19. <p align="center">
  20. <img src="Documentation/images/demo.gif" alt="LumaBrowser showing that a model fits this GPU with its predicted speed, then a local model searching the web and answering in chat" width="900">
  21. </p>
  22. ## Download
  23. Get the file for your computer from the
  24. **[latest release](https://github.com/amurgola/LumaBrowser/releases/latest)**
  25. (under "Assets"), then open it.
  26. | Your computer | File to download |
  27. |---|---|
  28. | Windows | `LumaBrowser.Setup.<version>.exe` (or the `portable` one, which needs no install) |
  29. | Windows on Arm (Snapdragon X) | `LumaBrowser.Setup.<version>-arm64.exe` |
  30. | Mac (Apple Silicon) | `LumaBrowser-<version>-arm64.dmg` |
  31. | Linux | `LumaBrowser-<version>.AppImage`, or `-arm64.AppImage` on Arm |
  32. Or, with Node 18 or newer, fetch and launch the latest build from a terminal:
  33. ```sh
  34. npx lumabrowser start
  35. ```
  36. ## One app instead of five
  37. Most local AI setups are a pile of separate tools. LumaBrowser is those tools
  38. as one desktop app that shares your GPUs, your models and your browser.
  39. | What people run today | In LumaBrowser |
  40. |---|---|
  41. | A model server (Ollama, LM Studio, llama.cpp) | Runtimes for every GPU vendor, with each model sized to your hardware before you download it |
  42. | A chat interface | Chat with web browsing, files, a knowledge base and your own agents |
  43. | An image tool | Image, video and music generation, plus area editing and outpainting |
  44. | A voice stack | Speech to text, text to speech and voice cloning |
  45. | A headless browser for agents | Your real browser, with your logins, over MCP, ACP, WebDriver and CDP |
  46. | A cloud computer-use service | Desktop control that runs locally: no per-step cost, and screenshots never leave the machine |
  47. ## First run
  48. **1. Pick a model.** Open **Models** and search. LumaBrowser reads your
  49. hardware and puts a fit badge and a predicted speed on every model before you
  50. download it. Models you already have in LM Studio, Ollama or the Hugging Face
  51. cache can be linked without downloading them again.
  52. <p align="center">
  53. <img src="Documentation/images/model-fit.png" alt="Model search showing that a quantization fits this GPU, with its predicted speed" width="800">
  54. </p>
  55. **2. Give it a task.** Open a chat and ask for something that needs the web:
  56. > Search the web for the latest llama.cpp release and tell me in a few bullets what changed.
  57. The agent opens its own tabs, reads the pages and answers. Your tabs stay yours.
  58. **3. Connect what you already use.** Point Claude Code, Codex, Cline, OpenCode
  59. or your editor at it from **Settings > Editors & agents**.
  60. ## What is inside
  61. | | |
  62. |---|---|
  63. | **[Browser](Documentation/features/browser.md)**<br>A Chromium browser with an ad blocker, Chrome extensions, a password manager, private windows, and a Shift Shift search for any setting or chat. | **[Local models](Documentation/features/local-models.md)**<br>llama.cpp and MLX runtimes for NVIDIA, AMD, Intel, Qualcomm and Apple hardware. Splits a model across GPUs or machines, and speeds it up with a drafter. |
  64. | **[Images, video, music and voice](Documentation/features/media.md)**<br>FLUX, Qwen-Image, MiniMax-H3, Wan 2.2 and more, a CivitAI browser, and edits that change only the area you paint. | **[Agents and coding](Documentation/features/agents.md)**<br>Named agents, a coding agent that also runs in your terminal and editor, tools the agent writes for itself, and signed-in sites turned into APIs. |
  65. | **[Computer use](Documentation/features/computer-use.md)**<br>The agent operates other apps on your desktop, shows which window it is in, and always asks before a password field or a checkout. | **[Automation](Documentation/features/automation.md) and [sharing](Documentation/features/sharing.md)**<br>Triggers, page watches and recurring tasks. Share your models and GPUs with your other devices, or a live tab by link. |
  66. What changed in each version is in the
  67. [release notes](https://github.com/amurgola/LumaBrowser/releases).
  68. ## For your agents and tools
  69. | You want to... | Use |
  70. |---|---|
  71. | Let Claude Code, Codex, Cline or OpenCode use a real browser | The MCP server, written into their config for you from Settings |
  72. | Run your LumaBrowser agents inside JetBrains Air or Zed | `luma acp` (Agent Client Protocol) |
  73. | Point Selenium, Puppeteer or Playwright at it | WebDriver on `127.0.0.1:9515`, CDP on `127.0.0.1:9222` (both off by default) |
  74. | Call local models from your own code | The OpenAI and Anthropic compatible API on `127.0.0.1:8317` (off by default) |
  75. | A coding agent in your terminal or editor | The `luma` CLI ([cli/README.md](cli/README.md)) and the VS Code and JetBrains plugins |
  76. Setup for each is in
  77. [Agents and coding](Documentation/features/agents.md#connecting-outside-agents),
  78. and the full API reference is at [lumabyte.com/apis](https://lumabyte.com/apis).
  79. ## Will it run on my computer?
  80. | Tier | Video memory | What runs well |
  81. |---|---|---|
  82. | Laptop | under 8 GB, or no GPU | Small LLMs, speech, voices; image generation needs 8 GB |
  83. | Gaming | 8 to 24 GB | Mid-size LLMs, image generation |
  84. | High-end | 24 GB or more, or a Mac with 32 GB+ | Large LLMs, video; music needs about 34 GB |
  85. Tiers use your largest single card, since cards do not add up for most
  86. workloads (LLMs are the exception: they can split across cards).
  87. | GPU | Support |
  88. |---|---|
  89. | NVIDIA | CUDA 12 and CUDA 13 (recommended for RTX 50) on Windows; Vulkan or CUDA on Linux |
  90. | AMD | Vulkan on Windows and Linux; ROCm when the AMD HIP SDK (Windows) or ROCm (Linux) is installed |
  91. | Intel | Vulkan on Windows and Linux; SYCL for Arc and Core Ultra graphics; OpenVINO (experimental) for Intel GPUs and NPUs |
  92. | Qualcomm Adreno (Windows on Arm) | OpenCL for LLMs |
  93. | Apple Silicon | MLX or llama.cpp (Metal) for LLMs; CPU for image generation |
  94. | None | CPU builds of every runtime |
  95. ## What leaves your machine
  96. Your browsing, chats, models, generated media and files stay on your machine.
  97. Nothing you type into a chat is sent anywhere unless you choose a cloud model.
  98. This is everything that does go out, and how to turn it off:
  99. | What | Sent to | Default | Turn off |
  100. |---|---|---|---|
  101. | Usage ping: version, platform, machine id, every 5 minutes | lumabyte.com | On | Settings > Privacy & permissions > Usage data |
  102. | Update check | lumabyte.com | On | Settings > About > Updates |
  103. | Extension catalog | lumabyte.com | Only when you open Extensions | |
  104. | Chrome extension downloads and update checks: extension id, Chromium version | Google (clients2.google.com) | Only when you install or update one | |
  105. | CivitAI searches, model pages, thumbnails and downloads (with your CivitAI key, if you set one) | civitai.com | Only when you open Browse CivitAI | |
  106. | Ad-block lists, runtimes, models | Ghostery, GitHub, Hugging Face and model hosts | When needed | |
  107. No feature depends on the usage ping.
  108. **Security.** Browser automation hands an agent your browser: anything an MCP,
  109. WebDriver or CDP client can see, the model behind it can see. Connect only
  110. agents you trust, keep those ports on localhost (the default), and set an API
  111. key in **Settings > API & MCP** before you expose the REST API on port 3000,
  112. which needs no key by default. Text from tools, web pages and windows is
  113. treated as content to read, never as commands, and saved passwords are never
  114. shown to a model.
  115. ## Docker and development
  116. <details>
  117. <summary>Run it on a server with Docker</summary>
  118. The Docker image runs the full app headless and serves the desktop in your
  119. browser over noVNC. It is meant for servers and agent hosts.
  120. ```sh
  121. LUMA_VNC_PASSWORD=change-me docker compose up -d
  122. ```
  123. - `http://localhost:6080`: the app, through noVNC.
  124. - `http://localhost:3000`: REST API and MCP proxy.
  125. In Docker the API accepts any address and needs no key by default. Do not
  126. publish port 3000 to a network you do not trust without setting an API key.
  127. GPU acceleration is not used in the container.
  128. </details>
  129. <details>
  130. <summary>Build from source</summary>
  131. You need Node 20 and a C++ toolchain (Visual Studio 2022 Build Tools on
  132. Windows) for the native SQLite module.
  133. ```sh
  134. npm install
  135. npm run dev # start the app in developer mode
  136. npm test # unit tests under Electron
  137. npm run test:e2e # end-to-end suites
  138. npm run build # installers for the current platform
  139. npm run shots:readme # retake this README's screenshots and demo clip from a blank-slate app
  140. npm run readme:check # word budget, links and images of this README
  141. ```
  142. The [Documentation](Documentation/index.md) folder is written for coding agents
  143. as much as for people. Every code file has a matching doc at
  144. `Documentation/<same path>.md`, and
  145. [Documentation/index.md](Documentation/index.md) lists them all in one line
  146. each. Point your agent at the index first, and have it update the doc whenever
  147. it changes the file.
  148. </details>
  149. ## Contributing
  150. Bug reports and pull requests are welcome. Read [CONTRIBUTING.md](CONTRIBUTING.md)
  151. first: contributions need a signed Contributor License Agreement.
  152. ## License
  153. LumaBrowser is free software under the
  154. [GNU Affero General Public License v3.0 or later](LICENSE). You may use it,
  155. study it, change it and redistribute it. If you distribute a modified version,
  156. or run one for other people over a network, you must publish your changes
  157. under the same license. Running unmodified official builds, including inside a
  158. company, needs nothing more. Organizations that want to keep their
  159. modifications private, combine LumaBrowser with proprietary software, or ship
  160. private extensions can obtain a
  161. [commercial license](COMMERCIAL-LICENSE.md) from Lumabyte, LLC.
  162. LumaBrowser is owned and maintained by Lumabyte, LLC, an Ohio company, which
  163. also runs [lumabrowser.com](https://lumabrowser.com) and
  164. [lumabyte.com](https://lumabyte.com). If you distribute a modified version,
  165. mark it as modified and do not present it as an official LumaBrowser release
  166. (AGPL section 7(c), see [LICENSE](LICENSE)). Third-party components keep their
  167. own licenses, listed in [THIRD-PARTY-LICENSES](THIRD-PARTY-LICENSES).

Ratings and comments

Sign in to rate this skill.

Sign in

Free to use. If it’s helped, pass it on.