# kyuz0/amd-strix-halo-toolboxes

Repository: https://github.com/kyuz0/amd-strix-halo-toolboxes
Canonical: https://ross.abutalabs.com/products/amd-strix-halo-toolboxes
Language: Python
License Family: other
Last push: 2026-08-26T15:06:20+00:00

## Health v2 (maintenance only)
Score: 62/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 99, release rhythm 35, longevity 28
- inputs: {"age_days": 401, "days_push": 7, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases, no_license
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1881, forks 189 (observed 2026-08-28T04:05:48.526631+00:00)

## What it is
A collection of pre-built Toolbx/Docker containers ('toolboxes') for running llama.cpp LLM inference on AMD Ryzen AI Max 'Strix Halo' integrated GPUs, with Vulkan and ROCm backend variants. It also includes host configuration guides, performance benchmarks, and a VRAM/memory planning estimator.

## Use cases
- run llama.cpp on AMD Strix Halo iGPU
- set up ROCm or Vulkan LLM inference on Ryzen AI Max
- benchmark llama.cpp performance on gfx1151
- estimate VRAM needs for local LLMs
- run local LLMs in containers on Fedora
- distributed LLM inference across Strix Halo machines

## When to choose
- you have an AMD Ryzen AI Max (Strix Halo) machine and want pre-tuned llama.cpp containers
- you want tested Vulkan/ROCm backend options with known-good kernel and firmware versions
- you want benchmarks and memory planning guidance for iGPU LLM inference

## When to avoid
- you have an NVIDIA GPU or non-Strix-Halo AMD hardware
- you want a polished GUI app rather than CLI containers
- you need production support or a formal license - it's a hobby project

## Facets
- artifact type: application
- maturity: active
- function: llm-inference, container-runtime, benchmarking, developer-tools
- domain: large-language-models, gpu-computing, developer-tools, self-hosted
- platform: cli
- tags: llama.cpp, strix-halo, rocm, vulkan, toolbx, amd-ryzen-ai-max, gfx1151, containers, local-llm, linux, docker, amd

## Member repositories
- kyuz0/amd-strix-halo-toolboxes (main) score 62

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:05:48.526631+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T03:13:58.273137+00:00, confidence not recorded.
  - readme: https://github.com/kyuz0/amd-strix-halo-toolboxes (fetched 2026-08-28T04:05:48.526631+00:00, sha 9e9e6c53d7ac)
- Data as of 2026-08-30T08:39:29.467469+00:00.