All glossary terms
AI & Generation

What is FLUX?

Also known as: FLUX.1, Black Forest Labs FLUX

Black Forest Labs' 12 billion parameter diffusion model released August 2024. Produces higher-fidelity generations than SDXL at higher GPU cost. Used by textile AI platforms for high-detail generation.

In detail

FLUX is the diffusion model family from Black Forest Labs (founded by ex-Stability AI researchers including the original Stable Diffusion author Robin Rombach). With 12B parameters in FLUX.1, it produces noticeably higher-fidelity outputs than SDXL at the cost of higher GPU memory and longer generation time. FLUX excels at fine detail, complex compositions, and prompt adherence — making it well-suited for high-detail textile work like intricate florals, detailed paisleys, and complex paneled engineered prints. The trade-off is cost: FLUX generations cost 2-4× more in GPU time than SDXL, so most textile AI platforms use FLUX as a 'pro' tier and SDXL as the default. FLUX has variants (Schnell for fast inference, Pro for highest quality, Dev for self-hosting), each optimizing different points on the quality-cost curve.

Example

A designer needs a complex paisley with fine line work and gradient transitions for a luxury silk scarf. SDXL produces the overall composition in 8 seconds but the line work is inconsistent. FLUX Pro produces the same composition in 22 seconds with significantly cleaner line work — worth the extra cost for this use case.

Related terms

Diffusion model
A class of generative AI models that produce images by iteratively denoising random Gaussian noise into coherent imagery. The dominant architecture for AI image generation in 2026, including textile pattern AI.
Stable Diffusion XL
Stability AI's open-source 2.6 billion parameter diffusion model released July 2023. The dominant model for textile pattern AI as of 2026, used by most production textile-AI platforms.
CFG scale
Classifier-free guidance scale — a parameter controlling how strictly a diffusion model follows the text prompt. Higher values produce more literal interpretations; lower values allow more creative variation.
Inpainting
An AI technique that fills in or modifies a masked region of an image while preserving the unmasked area. The standard method for AI-healing seamless tile boundaries.

Go deeper