Gentoo Packages

Search

Install

Install this package:

emerge -a sci-ml/fastflowlm

Package Information

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 an OpenAI-compatible server API for running language models entirely on the NPU. 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 CLI is MIT licensed; the NPU compute kernels are proprietary binaries.

Homepage:

https://fastflowlm.com/ https://github.com/ROCm/FastFlowLM

License:

MIT FastFlowLM-Terms

Versions

Metadata

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 an OpenAI-compatible server API for running language models entirely on the NPU. 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 CLI is MIT licensed; the NPU compute kernels are proprietary binaries.

Maintainers

• Fco. Javier Félix (web@inode64.com) - Type: person

Upstream

• Bugs: https://github.com/ROCm/FastFlowLM/issues
• Remote IDs:
  • github: ROCm/FastFlowLM

<pkgmetadata>
	<maintainer type="person" proxied="yes">
		<email>web@inode64.com</email>
		<name>Fco. Javier Félix</name>
	</maintainer>
	<longdescription lang="en">
        FastFlowLM (FLM) is a lightweight LLM inference runtime purpose-built for AMD Ryzen AI NPUs
        (XDNA2 architecture). It provides an Ollama-style CLI and an OpenAI-compatible server API for
        running language models entirely on the NPU. 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 CLI is MIT licensed; the NPU compute kernels are proprietary binaries.
    </longdescription>
	<upstream>
		<doc>https://fastflowlm.com/docs/</doc>
		<bugs-to>https://github.com/ROCm/FastFlowLM/issues</bugs-to>
		<remote-id type="github">ROCm/FastFlowLM</remote-id>
	</upstream>
</pkgmetadata>

Lint Warnings

• [Warning] Ebuild 9999 has a line exceeding 80 positions on line 85.
• [Warning] Ebuild 9999 has a line exceeding 80 positions on line 86.
• [Warning] Ebuild 9999 has a line exceeding 80 positions on line 121.
• [Warning] Ebuild 9999 has a line exceeding 80 positions on line 125.
• [Error] Ebuild sci-ml/fastflowlm/fastflowlm-9999.ebuild attempts to install into invalid path '/models' using 'keepdir'.
• Bypassing the network sandbox using `RESTRICT="network-sandbox"` violates Gentoo's core packaging policies. Ebuilds must support offline and reproducible builds, meaning all assets must be fetched via `SRC_URI` during the fetch phase and verified using Manifest checksums. To resolve this, you should package the required dependencies (e.g., node modules, cargo crates) into a tarball, host it, reference it in `SRC_URI` and perform an offline installation.
• [Error] Missing or unparsable Manifest file
• [Error] Package has a proxied maintainer but no proxy developer/project is listed

Files

• fastflowlm.confd
• fastflowlm.initd
• fastflowlm.service