sci-ml/amd-gaia::stuff
- Ebuilds: 3, Testing: 0.24.1 Description:
GAIA is AMD's open-source agent framework for local AI agents on
Ryzen AI hardware (NPU + iGPU). It orchestrates LLM-driven workflows
over any OpenAI-compatible inference endpoint, with built-in
integrations for Docker, Jira, code-search, RAG, MCP servers, and
Whisper / Kokoro voice pipelines. The reference local backend is
Lemonade Server (sci-ml/lemonade); GAIA itself is hardware-agnostic
so long as the upstream LLM API is OpenAI-compatible.
Homepage:https://github.com/amd/gaia License: MIT
sci-ml/fastflowlm::stuff
- Ebuilds: 4, Testing: 1.0.7, Snapshot: 9999 Description:
FastFlowLM (FLM) is a lightweight LLM inference runtime purpose-built
for AMD Ryzen AI NPUs (XDNA2 architecture). It provides an Ollama-style
CLI and OpenAI-compatible server API for running language models entirely
on the NPU with no GPU or CPU compute required.
Supported hardware: Ryzen AI 300-series (Strix Point, Strix Halo),
400-series (Gorgon Point), and Z2 Extreme. XDNA1 (Ryzen AI 7000/8000)
is NOT supported.
The orchestration code and CLI are MIT-licensed. NPU compute kernels
(xclbins) are proprietary binaries, free for commercial use under
$10M annual company revenue.
Homepage:
https://fastflowlm.com/
https://github.com/ROCm/FastFlowLM
License: MIT FastFlowLM-Terms
sci-ml/hf_xet::stuff
- Ebuilds: 1, Testing: 1.6.0 Description: xet client tech, used in huggingface_hub
Homepage:https://github.com/huggingface/xet-core License: Apache-2.0
Apache-2.0 Apache-2.0-with-LLVM-exceptions BSD-2 BSD CDDL
CDLA-Permissive-2.0 ISC MIT MPL-2.0 Unicode-3.0 ZLIB
sci-ml/kokoros::stuff
- Ebuilds: 1, Snapshot: 9999 Description:
Kokoros is a Rust implementation of the Kokoro-82M text-to-speech
model. Provides the `koko` CLI and an OpenAI-compatible HTTP server
used as the kokoro:cpu backend by sci-ml/lemonade.
Tracks upstream lucasjinreal/Kokoros directly. The lemonade-sdk
fork only diverges in CI infrastructure plus a bundled espeak-ng-data
copy that ::gentoo already provides via app-accessibility/espeak-ng,
so source-build users get the same binary either way.
Runtime model files (kokoro-v1.0.onnx + voices-v1.0.bin) are not
bundled — see pkg_postinst for a quick fetch recipe.
Homepage:https://github.com/lucasjinreal/Kokoros License: Apache-2.0
sci-ml/lemonade::stuff
- Ebuilds: 5, Testing: 2026.40.0, Snapshot: 9999 Description:
Lemonade is a local AI server that exposes optimized LLMs through
OpenAI / Anthropic / Ollama compatible APIs, running inference on
AMD NPU and GPU. The C++ server core (lemond, lemonade-server) is
packaged here without the Tauri desktop wrapper; the bundled web
frontend is available behind USE=webui, and the Tauri wrapper stays
CMake-toggleable and can be added behind a USE flag later if needed.
Pairs with sci-ml/fastflowlm to drive the AMD Ryzen AI XDNA2 NPU
backend.
Homepage:
https://lemonade-server.ai/
https://github.com/lemonade-sdk/lemonade
License: Apache-2.0
sci-ml/ollama::stuff
- Ebuilds: 3, Testing: 0.35.1 Description: Get up and running with Llama 3, Mistral, Gemma, and other language models
Homepage:https://ollama.com License: MIT
sci-ml/openflowlm::stuff
- Ebuilds: 1, Snapshot: 9999 Description:
OpenFlowLM is a community fork of FastFlowLM, the NPU-first LLM runtime
for AMD Ryzen AI (XDNA2) processors. For most of its LLM families it
replaces the closed NPU kernels with open AIE designs compiled from
source with the IRON/mlir-aie toolchain. Families without an open
design yet (vision, audio, Gemma 4, GPT-OSS, Whisper) still use
FastFlowLM's closed kernels.
Homepage:https://github.com/Atomic-Germ/OpenFlowLM-Next License: MIT Apache-2.0 FastFlowLM-Terms
sci-ml/sherpa-onnx::stuff
- Ebuilds: 3, Testing: 1.13.8 Description:
sherpa-onnx is a speech-stack toolkit from the k2-fsa project:
speech-to-text, text-to-speech, speaker diarization, voice activity
detection, source separation, and keyword spotting, all running on
ONNX Runtime (no PyTorch dependency).
Source build against system sci-libs/onnxruntime. For the prebuilt
-bin alternative (faster install, ships upstream's manylinux wheels)
see sci-ml/sherpa-onnx-bin.
The CMake build vendors a dozen small deps (eigen, asio, cargs, json,
kaldi-{decoder,native-fbank,fst}, openfst, kissfft, simple-sentencepiece,
hclust-cpp, optionally espeak-ng + piper-phonemize + portaudio +
websocketpp + pybind11) via FetchContent. The ebuild pre-fetches them
all via SRC_URI and stages into ${S} for the cmake fallback paths;
no network access during build.
Runtime model files for each task (ASR, diarization, TTS, etc.) live
upstream — see https://k2-fsa.github.io/sherpa/onnx/pretrained_models/
Homepage:
https://k2-fsa.github.io/sherpa/onnx/
https://github.com/k2-fsa/sherpa-onnx
License: Apache-2.0
sci-ml/sherpa-onnx-bin::stuff
- Ebuilds: 3, Testing: 1.13.8 Description:
sherpa-onnx is a speech-stack toolkit from the k2-fsa project:
speech-to-text, text-to-speech, speaker diarization, voice activity
detection, source separation, and keyword spotting, all running on
ONNX Runtime (no PyTorch dependency). Suited to CPU-only deployment
and embedded targets.
This -bin ebuild ships upstream's manylinux wheels (sherpa-onnx-core
for the C++ shared libraries plus a per-CPython-ABI wheel for the
Python bindings). Runtime model files are not bundled — see the
post-install message for download pointers.
Homepage:
https://k2-fsa.github.io/sherpa/onnx/
https://github.com/k2-fsa/sherpa-onnx
https://pypi.org/project/sherpa-onnx/
License: Apache-2.0
sci-ml/tokenizers::stuff
- Ebuilds: 3, Testing: 0.23.2 Description: Implementation of today's most used tokenizers
Homepage:https://github.com/huggingface/tokenizers License: Apache-2.0
Apache-2.0 Apache-2.0-with-LLVM-exceptions BSD BSD-2 CDLA-Permissive-2.0 ISC MIT MPL-2.0 Unicode-3.0 ZLIB
sci-ml/unsloth-desktop::stuff
- Ebuilds: 3, Testing: 0.1.902_beta Description: Tauri desktop UI to run and train AI models locally with Unsloth
Homepage:
https://unsloth.ai/docs/desktop
https://github.com/unslothai/unsloth
License: AGPL-3 OFL-1.1
0BSD Apache-2.0 BSD-2 BSD BlueOak-1.0.0 CC-BY-4.0 ISC MIT MPL-2.0
OFL-1.1 PYTHON Unlicense ZLIB
Apache-2.0 Apache-2.0-with-LLVM-exceptions BSD Boost-1.0 CC0-1.0
CDLA-Permissive-2.0 ISC MIT MPL-2.0 Unicode-3.0 ZLIB
sci-ml/unsloth-desktop-bin::stuff
- Ebuilds: 3, Testing: 0.1.902_beta Description:
Unsloth Desktop is a Tauri-based GUI shell for running and training LLMs,
diffusion, and audio models on local hardware. This binary package ships
only the desktop UI (unsloth-studio); on first launch it bootstraps a
private Python backend under ~/.unsloth/studio -- downloading uv, a
dedicated CPython, PyTorch and the inference engine -- independently of the
system sci-ml/unsloth stack. AMD ROCm is auto-selected on supported GPUs.
Homepage:
https://unsloth.ai/docs/desktop
https://github.com/unslothai/unsloth
License: AGPL-3