Skip to content
Back to search

lemonade

Lemonade helps users discover and run local AI apps by serving optimized LLMs right from their own GPUs and NPUs. Join our discord: https://discord.gg/5xXzkMu8Zk

Safety scan

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

Checked today

Check it deeper on your machine

It has changed since its scan on 11 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/lemonade-sdk/lemonade 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/lemonade-sdk/lemonade'

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/lemonade-sdk--lemonade/badge.svg)](/listings/lemonade-sdk--lemonade)

HTML

<a href="/listings/lemonade-sdk--lemonade"><img src="/api/listings/lemonade-sdk--lemonade/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 · 580 lines

View this file on GitHub (read at the repository's current version: the scan did not record which commit it read)

This file has changed since it was scanned, so this is not exactly the text that was scanned, and no line is marked. The findings are listed with the safety scan above.

  1. ## 🍋 Lemonade: Refreshingly fast local AI
  2. <p align="center">
  3. <a href="https://discord.gg/5xXzkMu8Zk">
  4. <img src="https://img.shields.io/badge/Discord-7289DA?logo=discord&logoColor=white" alt="Discord" /></a>
  5. <a href="https://github.com/lemonade-sdk/lemonade/blob/main/docs/dev/contribute.md" title="Contribution Guide">
  6. <img src="https://img.shields.io/badge/PRs-welcome-brightgreen.svg" alt="PRs Welcome" /></a>
  7. <a href="https://github.com/lemonade-sdk/lemonade/releases/latest" title="Download the latest release">
  8. <img src="https://img.shields.io/github/v/release/lemonade-sdk/lemonade?include_prereleases" alt="Latest Release" /></a>
  9. <a href="https://tooomm.github.io/github-release-stats/?username=lemonade-sdk&repository=lemonade">
  10. <img src="https://img.shields.io/github/downloads/lemonade-sdk/lemonade/total.svg" alt="GitHub downloads" /></a>
  11. <a href="https://github.com/lemonade-sdk/lemonade/issues">
  12. <img src="https://img.shields.io/github/issues/lemonade-sdk/lemonade" alt="GitHub issues" /></a>
  13. <a href="https://github.com/lemonade-sdk/lemonade/blob/main/LICENSE">
  14. <img src="https://img.shields.io/badge/License-Apache-yellow.svg" alt="License: Apache" /></a>
  15. <a href="https://star-history.com/#lemonade-sdk/lemonade">
  16. <img src="https://img.shields.io/badge/Star%20History-View-brightgreen" alt="Star History Chart" /></a>
  17. </p>
  18. <p align="center">
  19. <img src="https://github.com/lemonade-sdk/assets/blob/main/docs/banner_03.png?raw=true" alt="Lemonade Banner" />
  20. </p>
  21. <h3 align="center">
  22. <a href="./docs/guide/install/README.md">Download</a> |
  23. <a href="./docs/README.md">Documentation</a> |
  24. <a href="https://discord.gg/5xXzkMu8Zk">Discord</a>
  25. </h3>
  26. Lemonade is the local AI server that gives you the same capabilities as cloud APIs, except 100% free and private. Use the latest models for chat, coding, speech, and image generation on your own NPU and GPU.
  27. Lemonade comes in two flavors:
  28. * **Lemonade Server** installs a service you can connect to hundreds of great apps using standard OpenAI, Anthropic, and Ollama APIs.
  29. * **Embeddable Lemonade** is a portable binary you can package into your own application to give it multi-modal local AI that auto-optimizes for your user’s PC.
  30. *This project is built by the community for every PC, with optimizations by AMD engineers to get the most from Ryzen AI, Radeon, and Strix Halo PCs.*
  31. ## Getting Started
  32. 1. **Install**: [Windows](./docs/guide/install/windows.md) · [Linux](./docs/guide/install/linux.md) · [macOS](./docs/guide/install/macos.md) · [Docker](./docs/guide/install/docker.md) · [Source](./docs/dev/getting-started.md)
  33. 2. **Get Models**: Browse and download with the [Model Manager](#model-library)
  34. 3. **Generate**: Try models with the built-in interfaces for chat, image gen, speech gen, and more
  35. 4. **Mobile**: Take your lemonade to go: [iOS](https://apps.apple.com/us/app/lemonade-mobile/id6757372210) · [Android](https://play.google.com/store/apps/details?id=com.lemonade.mobile.chat.ai&pli=1) · [Source](https://github.com/lemonade-sdk/lemonade-mobile)
  36. 5. **Connect**: Use Lemonade with your [favorite apps](https://lemonade-server.ai/marketplace):
  37. <!-- MARKETPLACE_START -->
  38. <p align="center">
  39. <a href="./docs/integrations/claude-code.md" title="Claude Code"><img src="https://raw.githubusercontent.com/lemonade-sdk/marketplace/main/apps/claude-code/logo.png" alt="Claude Code" width="60" /></a>&nbsp;&nbsp;<a href="https://quickthoughts.ca/posts/firefox-chatback-lemonade-sdk/" title="Firefox Chatbot"><img src="https://raw.githubusercontent.com/lemonade-sdk/marketplace/main/apps/fx-chatbot/logo.png" alt="Firefox Chatbot" width="60" /></a>&nbsp;&nbsp;<a href="./docs/integrations/anythingLLM.md" title="AnythingLLM"><img src="https://raw.githubusercontent.com/lemonade-sdk/marketplace/main/apps/anythingllm/logo.png" alt="AnythingLLM" width="60" /></a>&nbsp;&nbsp;<a href="https://marketplace.dify.ai/plugins/langgenius/lemonade" title="Dify"><img src="https://raw.githubusercontent.com/lemonade-sdk/marketplace/main/apps/dify/logo.png" alt="Dify" width="60" /></a>&nbsp;&nbsp;<a href="https://github.com/amd/gaia" title="GAIA"><img src="https://raw.githubusercontent.com/lemonade-sdk/marketplace/main/apps/gaia/logo.png" alt="GAIA" width="60" /></a>&nbsp;&nbsp;<a href="https://admcpr.com/local-github-copilot-with-lemonade-server-on-windows" title="GitHub Copilot"><img src="https://raw.githubusercontent.com/lemonade-sdk/marketplace/main/apps/github-copilot/logo.png" alt="GitHub Copilot" width="60" /></a>&nbsp;&nbsp;<a href="https://github.com/lemonade-sdk/infinity-arcade" title="Infinity Arcade"><img src="https://raw.githubusercontent.com/lemonade-sdk/marketplace/main/apps/infinity-arcade/logo.png" alt="Infinity Arcade" width="60" /></a>&nbsp;&nbsp;<a href="https://n8n.io/integrations/lemonade-model/" title="n8n"><img src="https://raw.githubusercontent.com/lemonade-sdk/marketplace/main/apps/n8n/logo.png" alt="n8n" width="60" /></a>&nbsp;&nbsp;<a href="./docs/integrations/open-webui.md" title="Open WebUI"><img src="https://raw.githubusercontent.com/lemonade-sdk/marketplace/main/apps/open-webui/logo.png" alt="Open WebUI" width="60" /></a>&nbsp;&nbsp;<a href="./docs/integrations/open-hands.md" title="OpenHands"><img src="https://raw.githubusercontent.com/lemonade-sdk/marketplace/main/apps/openhands/logo.png" alt="OpenHands" width="60" /></a>
  40. </p>
  41. <p align="center"><em>Want your app featured here? <a href="https://github.com/lemonade-sdk/marketplace">Just submit a marketplace PR!</a></em></p>
  42. <!-- MARKETPLACE_END -->
  43. ## Supported Platforms
  44. | Platform | Build |
  45. |----------|-------|
  46. | [![Arch Linux](https://img.shields.io/badge/Arch%20Linux-supported-1793D1?logo=arch-linux&logoColor=white)](./docs/guide/install/arch.md) | [![Arch package](https://img.shields.io/archlinux/v/extra/x86_64/lemonade-server?label=Arch%20package)](https://archlinux.org/packages/extra/x86_64/lemonade-server) |
  47. | [![Debian Trixie+](https://img.shields.io/badge/Debian-Trixie%2B-A81D33?logo=debian&logoColor=white)](./docs/guide/install/debian.md) | [![Build .deb](https://img.shields.io/github/actions/workflow/status/lemonade-sdk/lemonade/cpp_server_build_test_release.yml?branch=main&label=Build%20.deb)](https://github.com/lemonade-sdk/lemonade/actions/workflows/cpp_server_build_test_release.yml) |
  48. | [![Docker](https://img.shields.io/badge/Docker-supported-2496ED?logo=docker&logoColor=white)](./docs/guide/install/docker.md) | [![Build Container Image](https://img.shields.io/github/actions/workflow/status/lemonade-sdk/lemonade/build-and-push-container.yml?branch=main&label=Build%20Container%20Image)](https://github.com/lemonade-sdk/lemonade/actions/workflows/build-and-push-container.yml) |
  49. | [![Fedora 43+](https://img.shields.io/badge/Fedora-43%2B-294172?logo=fedora&logoColor=white)](./docs/guide/install/fedora.md) | [![Build .rpm](https://img.shields.io/github/actions/workflow/status/lemonade-sdk/lemonade/cpp_server_build_test_release.yml?branch=main&label=Build%20.rpm)](https://github.com/lemonade-sdk/lemonade/actions/workflows/cpp_server_build_test_release.yml) |
  50. | [![macOS](https://img.shields.io/badge/macOS-supported-999999?logo=apple&logoColor=white)](./docs/guide/install/macos.md) | [![Build .pkg](https://img.shields.io/github/actions/workflow/status/lemonade-sdk/lemonade/cpp_server_build_test_release.yml?branch=main&label=Build%20.pkg)](https://github.com/lemonade-sdk/lemonade/actions/workflows/cpp_server_build_test_release.yml) |
  51. | [![Snap](https://img.shields.io/badge/Snap-supported-82BEA0?logo=snapcraft&logoColor=white)](https://snapcraft.io/lemonade-server) | [![Build Snap](https://img.shields.io/github/actions/workflow/status/lemonade-sdk/lemonade-server-snap/snap-build.yaml?branch=main&label=Build%20Snap)](https://github.com/lemonade-sdk/lemonade-server-snap/actions/workflows/snap-build.yaml) |
  52. | [![Ubuntu 24.04+](https://img.shields.io/badge/Ubuntu-24.04%2B-E95420?logo=ubuntu&logoColor=white)](./docs/guide/install/ubuntu.md) | [![Build Launchpad PPA](https://img.shields.io/github/actions/workflow/status/lemonade-sdk/lemonade/launchpad-ppa.yml?branch=main&label=Build%20Launchpad%20PPA)](https://github.com/lemonade-sdk/lemonade/actions/workflows/launchpad-ppa.yml) |
  53. | [![Windows 11](https://img.shields.io/badge/Windows-11-0078D6?logo=windows&logoColor=white)](./docs/guide/install/windows.md) | [![Build .msi](https://img.shields.io/github/actions/workflow/status/lemonade-sdk/lemonade/cpp_server_build_test_release.yml?branch=main&label=Build%20.msi)](https://github.com/lemonade-sdk/lemonade/actions/workflows/cpp_server_build_test_release.yml) |
  54. ## Using the CLI
  55. To run and chat with Gemma:
  56. ```
  57. lemonade run Gemma-4-E2B-it-GGUF
  58. ```
  59. To code with Lemonade models:
  60. ```
  61. lemonade launch claude
  62. ```
  63. Multi-modality:
  64. ```
  65. # image gen
  66. lemonade run SDXL-Turbo
  67. # speech gen
  68. lemonade run kokoro-v1
  69. # transcription
  70. lemonade run Whisper-Large-v3-Turbo
  71. ```
  72. To see available models and download them:
  73. ```
  74. lemonade list
  75. lemonade pull Gemma-4-E2B-it-GGUF
  76. ```
  77. To manage model aliases for environment-independent naming and active-standby failover:
  78. ```
  79. lemonade alias add production-llm Gemma-4-E2B-it-GGUF
  80. lemonade alias list
  81. # Instant active-standby failover to a different model target
  82. lemonade alias add production-llm Qwen3-0.6B-GGUF
  83. lemonade alias remove production-llm
  84. ```
  85. To see the backends available on your PC:
  86. ```
  87. lemonade backends
  88. ```
  89. For hybrid setups, Lemonade can also route to any OpenAI-compatible cloud provider (Fireworks, OpenAI, OpenRouter, Together, …) alongside local models — see [Cloud Offload](./docs/guide/configuration/cloud.md). *(Experimental.)*
  90. ## Model Library
  91. <img align="right" src="https://github.com/lemonade-sdk/assets/blob/main/docs/model_manager_03.png?raw=true" alt="Model Manager" width="280" />
  92. Lemonade supports a wide variety of LLMs (**GGUF**, **FLM**, and **ONNX**), whisper, stable diffusion, etc. models across CPU, GPU, and NPU.
  93. Use `lemonade pull` or the built-in **Model Manager** to download models. Custom GGUF/ONNX models can be pulled from Hugging Face or ModelScope, with their source retained for future updates.
  94. **[Browse the included model catalog →](https://lemonade-server.ai/models.html)**
  95. <br clear="right"/>
  96. ## Supported Configurations
  97. Lemonade supports multiple inference engines for LLM, speech, TTS, and image generation, and each has its own backend and hardware requirements.
  98. <!-- BEGIN GENERATED: backends-matrix -->
  99. <table>
  100. <thead>
  101. <tr>
  102. <th>Modality</th>
  103. <th>Engine</th>
  104. <th>Backend</th>
  105. <th>Device</th>
  106. <th>OS</th>
  107. </tr>
  108. </thead>
  109. <tbody>
  110. <tr>
  111. <td rowspan="16"><strong>Text generation</strong></td>
  112. <td rowspan="8"><code>llamacpp</code></td>
  113. <td><code>system</code></td>
  114. <td><code>x86_64</code>/ARM64 CPU, GPU</td>
  115. <td>Linux</td>
  116. </tr>
  117. <tr>
  118. <td><code>metal</code></td>
  119. <td>Apple Silicon GPU</td>
  120. <td>macOS</td>
  121. </tr>
  122. <tr>
  123. <td><code>cuda</code></td>
  124. <td>NVIDIA GPUs (Turing or newer)**</td>
  125. <td>Windows, Linux</td>
  126. </tr>
  127. <tr>
  128. <td><code>vulkan</code></td>
  129. <td><code>x86_64</code> CPU, AMD iGPU, AMD dGPU; ARM64 CPU/GPU (Linux)</td>
  130. <td>Windows, Linux</td>
  131. </tr>
  132. <tr>
  133. <td><code>rocm</code></td>
  134. <td>AMD GPUs supported by ROCm</td>
  135. <td>Windows, Linux</td>
  136. </tr>
  137. <tr>
  138. <td><code>cpu</code></td>
  139. <td><code>x86_64</code> CPU; ARM64 CPU (Linux)</td>
  140. <td>Windows, Linux</td>
  141. </tr>
  142. <tr>
  143. <td><code>nathanw</code>‡</td>
  144. <td>AMD Strix Halo (Vulkan performance fork)</td>
  145. <td>Linux</td>
  146. </tr>
  147. <tr>
  148. <td><code>halobox</code>‡</td>
  149. <td>AMD Strix Halo (Halo Box ROCm fork)</td>
  150. <td>Linux</td>
  151. </tr>
  152. <tr>
  153. <td rowspan="1"><code>llamacpp-hrx</code></td>
  154. <td><code>hrx</code>†</td>
  155. <td>AMD GPUs (gfx1100, gfx1151)</td>
  156. <td>Linux</td>
  157. </tr>
  158. <tr>
  159. <td rowspan="1"><code>flm</code></td>
  160. <td><code>npu</code></td>
  161. <td>XDNA2 NPU</td>
  162. <td>Windows, Linux</td>
  163. </tr>
  164. <tr>
  165. <td rowspan="1"><code>ryzenai-llm</code></td>
  166. <td><code>npu</code></td>
  167. <td>XDNA2 NPU</td>
  168. <td>Windows</td>
  169. </tr>
  170. <tr>
  171. <td rowspan="1"><code>vllm</code></td>
  172. <td><code>rocm</code>†</td>
  173. <td>Strix Halo iGPU (gfx1151)</td>
  174. <td>Linux</td>
  175. </tr>
  176. <tr>
  177. <td rowspan="1"><code>ds4</code></td>
  178. <td><code>rocm</code>‡</td>
  179. <td>Prebuilt ds4 for AMD Strix Halo</td>
  180. <td>Linux</td>
  181. </tr>
  182. <tr>
  183. <td rowspan="1"><code>rocmfpx</code></td>
  184. <td><code>rocm</code>‡</td>
  185. <td>AMD Strix Halo</td>
  186. <td>Linux</td>
  187. </tr>
  188. <tr>
  189. <td rowspan="1"><code>halogen</code></td>
  190. <td><code>rocm</code>‡</td>
  191. <td>AMD Strix Halo</td>
  192. <td>Linux</td>
  193. </tr>
  194. <tr>
  195. <td rowspan="1"><code>gufo</code></td>
  196. <td><code>rocm</code>‡</td>
  197. <td>AMD Strix Halo</td>
  198. <td>Linux</td>
  199. </tr>
  200. <tr>
  201. <td rowspan="6"><strong>Speech-to-text</strong></td>
  202. <td rowspan="5"><code>whispercpp</code></td>
  203. <td><code>npu</code></td>
  204. <td>XDNA2 NPU</td>
  205. <td>Windows</td>
  206. </tr>
  207. <tr>
  208. <td><code>metal</code></td>
  209. <td>Apple Silicon GPU</td>
  210. <td>macOS</td>
  211. </tr>
  212. <tr>
  213. <td><code>vulkan</code></td>
  214. <td><code>x86_64</code> CPU</td>
  215. <td>Windows, Linux</td>
  216. </tr>
  217. <tr>
  218. <td><code>rocm</code></td>
  219. <td>Supported AMD ROCm iGPU/dGPU families*</td>
  220. <td>Windows, Linux</td>
  221. </tr>
  222. <tr>
  223. <td><code>cpu</code></td>
  224. <td><code>x86_64</code> CPU</td>
  225. <td>Windows, Linux</td>
  226. </tr>
  227. <tr>
  228. <td rowspan="1"><code>moonshine</code></td>
  229. <td><code>cpu</code></td>
  230. <td><code>x86_64</code>/<code>arm64</code> CPU</td>
  231. <td>Windows, Linux, macOS</td>
  232. </tr>
  233. <tr>
  234. <td rowspan="4"><strong>Text-to-speech</strong></td>
  235. <td rowspan="2"><code>kokoro</code></td>
  236. <td><code>metal</code></td>
  237. <td>Apple Silicon GPU</td>
  238. <td>macOS</td>
  239. </tr>
  240. <tr>
  241. <td><code>cpu</code></td>
  242. <td><code>x86_64</code> CPU</td>
  243. <td>Windows, Linux</td>
  244. </tr>
  245. <tr>
  246. <td rowspan="2"><code>openmoss</code></td>
  247. <td><code>cuda</code>†</td>
  248. <td>NVIDIA GPUs</td>
  249. <td>Windows, Linux</td>
  250. </tr>
  251. <tr>
  252. <td><code>vulkan</code>†</td>
  253. <td>Vulkan-capable GPUs</td>
  254. <td>Windows, Linux</td>
  255. </tr>
  256. <tr>
  257. <td rowspan="6"><strong>Audio generation</strong></td>
  258. <td rowspan="3"><code>thinksound</code></td>
  259. <td><code>cuda</code>†</td>
  260. <td>NVIDIA GPUs</td>
  261. <td>Windows, Linux</td>
  262. </tr>
  263. <tr>
  264. <td><code>vulkan</code>†</td>
  265. <td>Vulkan-capable GPUs</td>
  266. <td>Windows, Linux</td>
  267. </tr>
  268. <tr>
  269. <td><code>rocm</code>†</td>
  270. <td>Supported AMD ROCm iGPU/dGPU families (ROCm via TheRock)</td>
  271. <td>Windows, Linux</td>
  272. </tr>
  273. <tr>
  274. <td rowspan="3"><code>acestep</code></td>
  275. <td><code>cuda</code>†</td>
  276. <td>NVIDIA GPUs</td>
  277. <td>Windows, Linux</td>
  278. </tr>
  279. <tr>
  280. <td><code>vulkan</code>†</td>
  281. <td>Vulkan-capable GPUs</td>
  282. <td>Windows, Linux</td>
  283. </tr>
  284. <tr>
  285. <td><code>rocm</code>†</td>
  286. <td>Supported AMD ROCm iGPU/dGPU families (ROCm via TheRock)</td>
  287. <td>Windows, Linux</td>
  288. </tr>
  289. <tr>
  290. <td rowspan="7"><strong>Image generation</strong></td>
  291. <td rowspan="5"><code>sd-cpp</code></td>
  292. <td><code>metal</code></td>
  293. <td>Apple Silicon GPU</td>
  294. <td>macOS</td>
  295. </tr>
  296. <tr>
  297. <td><code>cuda</code></td>
  298. <td>NVIDIA GPUs (Turing or newer)**</td>
  299. <td>Windows, Linux</td>
  300. </tr>
  301. <tr>
  302. <td><code>vulkan</code></td>
  303. <td>Vulkan-capable GPUs</td>
  304. <td>Windows, Linux</td>
  305. </tr>
  306. <tr>
  307. <td><code>rocm</code></td>
  308. <td>Supported AMD ROCm iGPU/dGPU families*</td>
  309. <td>Windows, Linux</td>
  310. </tr>
  311. <tr>
  312. <td><code>cpu</code></td>
  313. <td><code>x86_64</code> CPU</td>
  314. <td>Windows, Linux</td>
  315. </tr>
  316. <tr>
  317. <td rowspan="2"><code>thenoise</code></td>
  318. <td><code>rocm</code></td>
  319. <td>Supported AMD ROCm families</td>
  320. <td>Windows, Linux</td>
  321. </tr>
  322. <tr>
  323. <td><code>cuda</code>†</td>
  324. <td>NVIDIA GPUs</td>
  325. <td>Linux</td>
  326. </tr>
  327. <tr>
  328. <td rowspan="4"><strong>3D generation</strong></td>
  329. <td rowspan="4"><code>trellis</code></td>
  330. <td><code>cuda12</code>†</td>
  331. <td>NVIDIA Pascal/Volta GPUs (CUDA 12)</td>
  332. <td>Windows, Linux</td>
  333. </tr>
  334. <tr>
  335. <td><code>cuda</code>†</td>
  336. <td>NVIDIA GPUs</td>
  337. <td>Windows, Linux</td>
  338. </tr>
  339. <tr>
  340. <td><code>vulkan</code>†</td>
  341. <td>Vulkan-capable GPUs</td>
  342. <td>Windows, Linux</td>
  343. </tr>
  344. <tr>
  345. <td><code>rocm</code>†</td>
  346. <td>Supported AMD ROCm iGPU/dGPU families (ROCm via TheRock)</td>
  347. <td>Windows, Linux</td>
  348. </tr>
  349. <tr>
  350. <td rowspan="3"><strong>Text classification</strong></td>
  351. <td rowspan="3"><code>onnxruntime</code></td>
  352. <td><code>cpu</code>†</td>
  353. <td><code>x86_64</code> CPU</td>
  354. <td>Windows</td>
  355. </tr>
  356. <tr>
  357. <td><code>cpu</code>†</td>
  358. <td><code>x86_64</code>/<code>arm64</code> CPU</td>
  359. <td>Linux</td>
  360. </tr>
  361. <tr>
  362. <td><code>cpu</code>†</td>
  363. <td><code>arm64</code> CPU</td>
  364. <td>macOS</td>
  365. </tr>
  366. </tbody>
  367. </table>
  368. † experimental backend, ‡ guest backend. See [Backend Tiers and Formats](./docs/api/lemonade.md#backend-tiers-and-formats).
  369. <!-- END GENERATED: backends-matrix -->
  370. To check exactly which recipes/backends are supported on your own machine, run:
  371. ```
  372. lemonade backends
  373. ```
  374. <details>
  375. <summary><small><i>* See supported AMD ROCm platforms</i></small></summary>
  376. <br>
  377. <table>
  378. <thead>
  379. <tr>
  380. <th>Architecture</th>
  381. <th>Platform Support</th>
  382. <th>GPU Models</th>
  383. </tr>
  384. </thead>
  385. <tbody>
  386. <tr>
  387. <td><b>gfx1151</b> (STX Halo)</td>
  388. <td>Windows, Ubuntu</td>
  389. <td>Ryzen AI MAX+ Pro 395</td>
  390. </tr>
  391. <tr>
  392. <td><b>gfx120X</b> (RDNA4)</td>
  393. <td>Windows, Ubuntu</td>
  394. <td>Radeon AI PRO R9700, RX 9070 XT/GRE/9070, RX 9060 XT</td>
  395. </tr>
  396. <tr>
  397. <td><b>gfx110X</b> (RDNA3)</td>
  398. <td>Windows, Ubuntu</td>
  399. <td>Radeon PRO W7900/W7800/W7700/V710, RX 7900 XTX/XT/GRE, RX 7800 XT, RX 7700 XT</td>
  400. </tr>
  401. </tbody>
  402. </table>
  403. </details>
  404. <details>
  405. <summary><small><i>** See supported NVIDIA CUDA platforms</i></small></summary>
  406. <br>
  407. <table>
  408. <thead>
  409. <tr>
  410. <th>Compute Capability</th>
  411. <th>Architecture</th>
  412. <th>GPU Models</th>
  413. </tr>
  414. </thead>
  415. <tbody>
  416. <tr>
  417. <td><b>sm_75</b></td>
  418. <td>Turing</td>
  419. <td>RTX 20-series, GTX 16-series, T4</td>
  420. </tr>
  421. <tr>
  422. <td><b>sm_80</b> / <b>sm_86</b></td>
  423. <td>Ampere</td>
  424. <td>RTX 30-series, A100, A40</td>
  425. </tr>
  426. <tr>
  427. <td><b>sm_89</b></td>
  428. <td>Ada Lovelace</td>
  429. <td>RTX 40-series, L40, L4</td>
  430. </tr>
  431. <tr>
  432. <td><b>sm_90</b></td>
  433. <td>Hopper</td>
  434. <td>H100, H200</td>
  435. </tr>
  436. <tr>
  437. <td><b>sm_100</b> / <b>sm_120</b></td>
  438. <td>Blackwell</td>
  439. <td>RTX 50-series, B100, B200</td>
  440. </tr>
  441. </tbody>
  442. </table>
  443. </details>
  444. ## Project Roadmap
  445. Lemonade's roadmap is defined by a set of working groups. Visit the landing page [here](./docs/dev/working-groups/README.md) to learn each group's goal and roadmap.
  446. ## Integrate Embeddable Lemonade in Your Application
  447. Embeddable Lemonade is a binary version of Lemonade that you can bundle into your own app to give it a portable, auto-optimizing, multi-modal local AI stack. This lets users focus on your app, with zero Lemonade installers, branding, or telemetry.
  448. Check out the [Embeddable Lemonade guide](docs/embeddable/README.md).
  449. ## Connect Lemonade Server to Your Application
  450. You can use any OpenAI-compatible client library by configuring it to use `http://localhost:13305/v1` as the base URL. A table containing official and popular OpenAI clients on different languages is shown below.
  451. Feel free to pick and choose your preferred language.
  452. | Python | C++ | Java | C# | Node.js | Go | Ruby | Rust | PHP |
  453. |--------|-----|------|----|---------|----|-------|------|-----|
  454. | [openai-python](https://github.com/openai/openai-python) | [openai-cpp](https://github.com/olrea/openai-cpp) | [openai-java](https://github.com/openai/openai-java) | [openai-dotnet](https://github.com/openai/openai-dotnet) | [openai-node](https://github.com/openai/openai-node) | [go-openai](https://github.com/sashabaranov/go-openai) | [ruby-openai](https://github.com/alexrudall/ruby-openai) | [async-openai](https://github.com/64bit/async-openai) | [openai-php](https://github.com/openai-php/client) |
  455. ### Python Client Example
  456. ```python
  457. from openai import OpenAI
  458. # Initialize the client to use Lemonade Server
  459. client = OpenAI(
  460. base_url="http://localhost:13305/api/v1",
  461. api_key="lemonade" # required but unused
  462. )
  463. # Create a chat completion
  464. completion = client.chat.completions.create(
  465. model="Gemma-4-E2B-it-GGUF", # or any other available model
  466. messages=[
  467. {"role": "user", "content": "What is the capital of France?"}
  468. ]
  469. )
  470. # Print the response
  471. print(completion.choices[0].message.content)
  472. ```
  473. Click to learn more about the [available APIs](./docs/api/README.md) and how to [embed Lemonade](./docs/embeddable/README.md) in your own application.
  474. ## FAQ
  475. To read our frequently asked questions, see our [FAQ Guide](./docs/guide/faq.md)
  476. ## Contributing
  477. Lemonade is built by the local AI community! If you would like to contribute to this project, please check out our [contribution guide](./docs/dev/contribute.md).
  478. ## Maintainers
  479. This is a community project with many maintainers, please see the [maintainers list here](./docs/dev/contribute.md#maintainers) to see their subject areas. You can reach us by filing an [issue](https://github.com/lemonade-sdk/lemonade/issues) or joining our [Discord](https://discord.gg/5xXzkMu8Zk). This project is sponsored by [AMD](mailto:lemonade@amd.com).
  480. ## Code Signing Policy
  481. Free code signing provided by [SignPath.io](https://signpath.io), certificate by [SignPath Foundation](https://signpath.org).
  482. - **Committers and reviewers**: [Maintainers](#maintainers) of this repo
  483. - **Approvers**: [Owners](https://github.com/orgs/lemonade-sdk/people?query=role%3Aowner)
  484. **Privacy policy**: This program will not transfer any information to other networked systems unless specifically requested by the user or the person installing or operating it. When the user requests a model download or registry lookup, Lemonade may contact [Hugging Face Hub](https://huggingface.co/) (see their [privacy policy](https://huggingface.co/privacy)) or [ModelScope](https://modelscope.cn/), according to the model source selected by the user or packager.
  485. ## License and Attribution
  486. This project is:
  487. - Built with C++ (server) and React (app) with ❤️ for the open source community,
  488. - Standing on the shoulders of great tools from:
  489. - [ggml/llama.cpp](https://github.com/ggml-org/llama.cpp)
  490. - [ggml/whisper.cpp](https://github.com/ggerganov/whisper.cpp)
  491. - [ggml/stable-diffusion.cpp](https://github.com/leejet/stable-diffusion.cpp)
  492. - [kokoros](https://github.com/lucasjinreal/Kokoros)
  493. - [OnnxRuntime GenAI](https://github.com/microsoft/onnxruntime-genai)
  494. - [Hugging Face Hub](https://github.com/huggingface/huggingface_hub)
  495. - [ModelScope](https://github.com/modelscope/modelscope)
  496. - [OpenAI API](https://github.com/openai/openai-python)
  497. - [IRON/MLIR-AIE](https://github.com/Xilinx/mlir-aie)
  498. - and more...
  499. - Licensed under the [Apache 2.0 License](https://github.com/lemonade-sdk/lemonade/blob/main/LICENSE).
  500. - Portions of the project are licensed as described in [LICENSE](./LICENSE).
  501. <!--This file was originally licensed under Apache 2.0. It has been modified.
  502. Modifications Copyright (c) 2025 AMD-->

Ratings and comments

Sign in to rate this skill.

Sign in

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