Skip to content
local-ai

FLUX.1 [dev]

Black Forest Labs · Image generation · 12B · Released 1 August 2024

Non-commercial only TextImage Open weights Apple Silicon

FLUX.1 dev is the higher-quality tier of Black Forest Labs' FLUX family, producing photorealistic images with strong prompt adherence and genuinely readable text, a long-standing weakness of earlier open models. It is a large 12B model, so it wants a capable GPU, and it runs across ComfyUI and InvokeAI with a growing library of LoRAs. The catch is licensing: dev is released under a non-commercial licence, so any commercial use needs a separate agreement with Black Forest Labs. If commercial use matters, the Apache-2.0 FLUX.1 schnell is the sibling to reach for, trading some quality for freedom and speed.

Strengths

  • Excellent image quality and prompt adherence
  • Renders readable text well
  • A large and growing ecosystem of fine-tunes and LoRAs

Weaknesses

  • Non-commercial licence, so you cannot ship commercial output without a licence
  • Slower than [schnell], needing more steps per image
  • A 12B model needs 16GB or more, or quantisation, for comfortable use

Hardware requirements

QuantisationApprox. VRAMNotes
GGUF Q4~8GBQuantised with offloading, fits an 8GB to 12GB card
FP8~12GBA good balance on a 16GB card
FP16~24GBFull precision, including the text encoder

Optimised builds available for Apple Silicon.

What you'd need to run this

Roughly what a machine to run this would need, at up to three levels of quality. Memory is the deciding factor.

Minimum to run it

GGUF Q4 · ~8GB needed

One 12GB GPU

NVIDIA GeForce RTX 3060 12GB

or a Mac or mini-PC with unified memory, if you prefer no discrete GPU, Mac mini M5 Pro .

At least 32GB of system RAM alongside the card.

around £700–£1,100

What else 12GB runs →

For good quality

FP8 · ~12GB needed

One 16GB GPU

Intel Arc A770 16GB

or a Mac or mini-PC with unified memory, if you prefer no discrete GPU, Mac mini M5 Pro .

At least 32GB of system RAM alongside the card.

around £700–£1,100

What else 16GB runs →

Best quality

FP16 · ~24GB needed

One 32GB GPU

NVIDIA GeForce RTX 5090

or a Mac or mini-PC with unified memory, if you prefer no discrete GPU, Mac mini M4 Pro .

64GB of system RAM alongside the card.

Licence

FLUX.1 [dev] Non-Commercial License read the licence

FLUX.1 [dev]: common questions

What hardware do I need to run FLUX.1 [dev]?
At its most compressed (GGUF Q4) it needs roughly 8GB of VRAM, and about 24GB for good quality. VRAM figures are approximate and depend on context length and settings.
Is FLUX.1 [dev] free for commercial use?
No. FLUX.1 [dev] is licensed for non-commercial use only. FLUX.1 [dev] is licensed for non-commercial use only.
Can I run FLUX.1 [dev] on Apple Silicon?
Yes. FLUX.1 [dev] has builds optimised for Apple Silicon, through MLX or GGUF on a Mac.

Availability

Recommended for

  • Personal and research image generation at high quality
  • Users who want the deepest FLUX fine-tune ecosystem
  • Evaluation before deciding whether to license commercially

Related models

Run it with

Related guides

Glossary

Our coverage

Catalogue entry last verified 30 July 2026. Specifications change; verify anything you are about to spend money on.