AMD Releases Lemonade 11.0 Local AI Server With Text-To-Speech, Other New Features
Lemonade remains a great way to try out the potential of accelerated AI locally across AMD hardware on both Windows and Linux. Beyond being a great AMD AI hardware demonstrator, it's useful in its own right while being fully open-source and supporting the latest large language models.
Lemonade 11.0 brings text-to-speech (TTS) support using an OpenMOSS back-end with support for voice cloning and voice-design models. This TTS support is integrated into the Lemonade UI.
Lemonade 11.0 also introduces a new 3D generation modality, complete with an in-app model viewer. Another nice improvement for Linux users with Lemonade 11.0 is the FastFlowLM NPU back-end now being automatically installed on Linux.
Highlights
- A new 3D-generation modality ports the Trellis.2 image-to-3D pipeline, exposing a POST /v1/3d/generations endpoint and a GUI 3D panel with an in-app model viewer.
- Text-to-speech arrives with the OpenMOSS backend, including voice-cloning and voice-design models and a dedicated TTS panel in the GUI.
- Naming a collection.router model on /chat/completions, /completions, or /responses now runs its routing engine and reports the selected route via an x_lemonade_route body field and x-lemonade-route header.
- ModelScope is now supported as a second remote model registry alongside Hugging Face, with source selection across the CLI, API, and desktop app.
- The FastFlowLM NPU backend now auto-installs on Linux instead of requiring a manual system-package install, with a new prefer_system config flag.
Downloads and more details on Lemonade 11.0 at GitHub.
14 Comments