skills.rest
mflux-community
mflux-communityOFFICIAL
Explore all skills by this author →
WebsiteGitHub

mflux-cli

Navigate MFLUX CLI commands and inventory supported image generation features.

2.4k|187|Updated Aug 10, 2024
Search Tags:#cli#image generation#Flux#mlx#LoRA#text-to-image#MFLUX
One-click install
npx skills add https://github.com/mflux-community/mflux --skill mflux-cli-mflux-community
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: mflux-cli
Source: https://github.com/mflux-community/mflux/tree/main/.cursor/skills/mflux-cli
Command: npx skills add https://github.com/mflux-community/mflux --skill mflux-cli-mflux-community

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? The MFLUX repository exposes many model-specific CLI entrypoints whose flags and defaults drift from README examples, making it hard to know which commands exist and how to invoke them correctly. ## Core Features & Use Cases - Capability Inventory: Lists supported CLI features across text-to-image, image-to-image, editing, upscaling, LoRA, and training. - Command Location Guidance: Explains where shared parsers, model-specific entrypoints, and utility CLIs live without relying on brittle file paths. - Best-Practice Defaults: Enforces correct inference step counts per model (e.g., turbo/lite variants) and flags like --low-ram and --stepwise-image-output-dir. - Use Case: A user asks which model to use for fast image generation; the skill points to z-image-turbo with its recommended 9-step default instead of base-model step counts. ## Quick Start Ask the AI to list the MFLUX CLI commands available for generating an image with z-image-turbo including recommended steps and quantization flags.

Frequently Asked Questions about mflux-cli

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I generate an image with the MFLUX CLI?▼

Use a model-specific entrypoint such as mflux-generate-z-image-turbo with --prompt, --width, --height, --seed, --steps, and -q for quantization. Check MODEL_INFERENCE_STEPS for the recommended step count before constructing the call.

What image generation models does MFLUX support?▼

MFLUX supports text-to-image across Flux, Flux2, Qwen, Z-Image Turbo, and FIBO, plus image-to-image where supported. It also includes Kontext conditioning, ControlNet, inpainting, Redux, and SeedVR2 upscaling.

How many inference steps should I use for turbo models?▼

Distilled variants like z-image-turbo and fibo-lite have low recommended step counts defined in MODEL_INFERENCE_STEPS, such as 9 steps for z-image-turbo. Using full base-model step counts is usually counterproductive for these models.

Does MFLUX support LoRA loading and training?▼

Yes, MFLUX supports loading single or multiple LoRAs with scale control, in-context style LoRA shortcuts, and a LoRA library lookup tool. It also provides a CLI for LoRA finetuning.

How can I reduce memory usage when running MFLUX?▼

Pass the --low-ram flag to reduce memory usage during generation. You can also use quantized inference with the -q flag, for example -q 8, to lower the model's memory footprint.

Related Skills

Explore More Software Engineering Skills→
89jobrien89jobrien

mlx-fine-tuning

Fine-tune LLMs with LoRA on Apple Silicon using MLX.

Community
Advanced
Pixazo-AIPixazo-AI

seedvr

Upscale and enhance images and videos using the SeedVR API via Pixazo.

Official
Intermediate
calesthiocalesthio
123

bria-fibo-image

Build rights-aware Bria FIBO image-generation workflows with validation and governance.

Community
Advanced
lxt070409lxt070409
1

llama-cpp

Run local inference with llama.cpp models and discover GGUF models from the Hugging Face Hub.

Community
Advanced