# mflux-community/mflux

Apple MLX native implementations of state-of-the-art generative image & video models

Repository: https://github.com/mflux-community/mflux
Canonical: https://ross.abutalabs.com/products/mflux
Language: Python
License: MIT
License Family: permissive
Topics: ai, apple-silicon, diffusers, flux, huggingface, ml, mlx, transformers, qwen, qwen-image, seedvr2, z-image, fibo
Last push: 2026-08-26T05:25:09+00:00

## Health v2 (maintenance only)
Score: 90/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 99, release rhythm 99, longevity 53
- inputs: {"age_days": 753, "days_push": 7, "days_rel": 8, "gap_med": 5, "n_releases_24m": 58}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 2291, forks 172 (observed 2026-08-28T04:06:34.931950+00:00)

## What it is
MFLUX is a native MLX implementation of state-of-the-art generative image and video models (Flux, Qwen-Image, Z-Image, and others), ported line-by-line from Hugging Face Diffusers/Transformers. It runs locally on Apple Silicon Macs via a CLI and Python API.

## Use cases
- generate images locally on a mac
- run flux text-to-image on apple silicon
- convert huggingface diffusers models to mlx
- generate images from a text prompt via cli
- run generative image models without a gpu server
- quantize image models for faster local inference

## When to choose
- you have an Apple Silicon Mac and want fast local image generation
- you want readable, minimal MLX implementations of diffusion models
- you prefer a simple CLI or small Python API over heavyweight frameworks

## When to avoid
- you need CUDA/NVIDIA GPU support
- you need the full flexibility of Hugging Face Diffusers
- you need production serving of image models at scale

## Facets
- artifact type: library
- maturity: active
- function: machine-learning, image-processing, cli, llm-inference
- domain: artificial-intelligence, image-processing, machine-learning, gpu-computing
- platform: python, cli
- tags: mlx, apple-silicon, diffusion-models, text-to-image, flux, generative-ai, local-inference, macos

## Member repositories
- mflux-community/mflux (main) score 90

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:06:34.931950+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 2026-08-30T02:40:40.971131+00:00, confidence not recorded.
  - readme: https://github.com/mflux-community/mflux (fetched 2026-08-28T04:06:34.931950+00:00, sha 83ce66d7fbe6)
- Data as of 2026-08-30T08:39:29.467469+00:00.