mflux

TLDR

Generate an image from a text prompt

$ mflux-generate --prompt "[a sunset over mountains]" --model [flux]

Specify image dimensions

$ mflux-generate --prompt "[prompt]" --width [1024] --height [768] --model [flux]

Set steps and seed

$ mflux-generate --prompt "[prompt]" --steps [20] --seed [42] --model [flux]

Save to specific path

$ mflux-generate --prompt "[prompt]" --output [output.png] --model [flux]

Use quantized model for lower memory

$ mflux-generate --prompt "[prompt]" --quantize [4] --model [flux]

Apply a LoRA adapter

$ mflux-generate --prompt "[prompt]" --lora-path [adapter.safetensors] --model [flux]

SYNOPSIS

mflux-generate --prompt text --model name [options]

mflux is a Flux image generation tool built natively on Apple's MLX framework, optimized for Apple Silicon (M1/M2/M3/M4). It generates images locally using Flux models without requiring a GPU server or cloud API.
Installation is via pip (`pip install mflux`). The tool downloads model weights on first use and caches them locally.
Flux-schnell is faster with fewer steps needed. Flux-dev produces higher quality but takes longer. Quantization (4-bit or 8-bit) reduces memory usage for running on machines with limited unified memory.
LoRA adapters allow fine-tuned styles and concepts to be applied on top of base models.

PARAMETERS

--prompt TEXT

Text prompt for image generation.

--model NAME

Model to use (flux, flux-dev, flux-schnell).

--output FILE

Output image path.

--width PX

Image width in pixels.

--height PX

Image height in pixels.

--steps N

Number of inference steps.

--seed INT

Random seed for reproducibility.

--quantize BITS

Quantization level (4 or 8 bit).

--lora-path FILE

Path to LoRA adapter weights.

--guidance FLOAT

Guidance scale.

CAVEATS

Apple Silicon only (M-series Macs). Requires Python 3.10+. Model downloads are several GB. Memory usage depends on model and quantization level. Limited to Flux model family.

HISTORY

mflux was created by Filip Strand in 2024 to bring native Flux image generation to Apple Silicon using the MLX framework. It provides a streamlined alternative to running Stable Diffusion through compatibility layers on macOS.