Gentoo Packages

sci-ml/fastflowlm (stuff)

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 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-NPU-Binary

Versions

Version EAPI Keywords Slot
9999 8 0
1.0.9 8 ~amd64 0
1.0.11 8 ~amd64 0
1.0.10 8 ~amd64 0

Metadata

Description

Maintainers

Upstream

Raw Metadata XML 
<pkgmetadata>
	<maintainer type="person">
		<email>iohann.s.titov@gmail.com</email>
		<name>Ivan S. Titov</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 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.
  </longdescription>
	<use>
		<flag name="gemma4e-source">Build the Gemma 4 E2B/E4B NPU engine from the source upstream opened in 1.0.8 instead of installing the prebuilt library; the other engines stay prebuilt</flag>
		<flag name="hrx">Use the HRX amdxdna runtime (<pkg>dev-libs/hrx-amdxdna-bin</pkg>) instead of XRT to drive the NPU; an upstream preview</flag>
		<flag name="openrc">Install OpenRC init.d/conf.d files for a supervise-daemon-managed service</flag>
		<flag name="systemd">Install a systemd template service unit (fastflowlm@.service)</flag>
	</use>
	<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

USE Flags

Manage flags for this package: euse -i <flag> -p sci-ml/fastflowlm | euse -E <flag> -p sci-ml/fastflowlm | euse -D <flag> -p sci-ml/fastflowlm

Flag Description 9999 1.0.9 1.0.11 1.0.10
gemma4e-source Build the Gemma 4 E2B/E4B NPU engine from the source upstream opened in 1.0.8 instead of installing the prebuilt library; the other engines stay prebuilt ✓ ✓ ✓ ✓
hrx Use the HRX amdxdna runtime (<pkg>dev-libs/hrx-amdxdna-bin</pkg>) instead of XRT to drive the NPU; an upstream preview ✓ ✓ ✓ ✓
openrc Install OpenRC init.d/conf.d files for a supervise-daemon-managed service ✓ ✓ ✓ ✓
systemd Install a systemd template service unit (fastflowlm@.service) ✓ ✓ ✓ ✓

Files

Manifest

Type File Size Versions
DIST fastflowlm-1.0.10.tar.gz 160697902 bytes 1.0.10
DIST fastflowlm-1.0.11.tar.gz 171486289 bytes 1.0.11
DIST fastflowlm-1.0.9.tar.gz 166808860 bytes 1.0.9
DIST msgpack-c-092bc69b6e815980bce7808595c914dd3a29f905.tar.gz 476163 bytes 1.0.9, 1.0.11, 1.0.10
DIST sentencepiece-11051e3b73b3a6222a52acd720e39805dc7545ab.tar.gz 13487870 bytes 1.0.9, 1.0.11, 1.0.10
DIST tokenizers-cpp-acbdc5a27ae01ba74cda756f94da698d40f11dfe.tar.gz 39758 bytes 1.0.9, 1.0.11, 1.0.10
DIST xgrammar-97787376faee5ed8466cfad57c99855e4ce2f6aa.tar.gz 1270123 bytes 1.0.11, 1.0.10
DIST xgrammar-dlpack-bbd2f4d32427e548797929af08cfe2a9cbb3cf12.tar.gz 190029 bytes 1.0.11, 1.0.10
Unmatched Entries
Type File Size