Directory/Image Generation/Image Generation/Black Forest Labs FLUX
Black Forest Labs FLUX

Black Forest Labs FLUX

Black Forest Labs (BFL) provides FLUX, a suite of production-grade diffusion models for high-fidelity text-to-image generation and advanced image editing. Its flagship FLUX.2 family offers multi-reference image input, supporting up to 10 images for precise control over composition, style, and subjects. Models range from the high-fidelity FLUX.2 [max] to ultra-fast, compact 4B/9B parameter 'Klein' variants optimized for real-time applications. Core capabilities include prompt-free outpainting, object removal (Eraser), and virtual try-on (VTO). The platform serves developers and businesses through a pay-per-use API with megapixel-based billing, open-weight licenses for self-hosting, and enterprise support. Target users include product teams needing scalable visual content, agencies creating marketing assets, and developers integrating AI image generation into applications. The models are known for exceptional prompt adherence and photorealistic output up to 4MP resolution.

Image Generation
Visit website
Added on
Jul 13, 2026
FLUX.2Multi-ref editingUp to 4MP9B/4B paramsOutpaintingObject removalAPI accessOpen weights
Product information

Everything worth knowing about Black Forest Labs FLUX

The complete picture — from everyday features to the technical detail builders and IT folks dig for.

Core features

What it actually does

Text-to-Image Generation
Generate photorealistic images from text prompts with high visual fidelity.
Multi-Reference Image Editing
Edit images using up to 10 reference images simultaneously for precise control over colors, poses, and composition.
Outpainting
Extend images beyond their original borders with coherent scene reconstruction, no prompt required.
Object Removal
Erase unwanted elements from images with automatic coherent filling.
Flexible Megapixel Billing
Pay only for the resolution you generate with no per-image minimums or subscription fees.
Technical capabilities
Transformer-Based Diffusion Architecture
FLUX models replace traditional U-Net with a transformer backbone for improved scaling and performance.
Compact Model Variants
Available in 4B and 9B parameter configurations optimized for different latency and throughput requirements.
Must watch videos
Martin Scorsese x Black Forest Labs
3:10
Tutorial3:10
Martin Scorsese x Black Forest Labs
Martin Scorsese has spent six decades bringing stories to life. He's joined Black Forest Labs as an advisor, helping shape visual intelligence for creators obsessed with their craft. "There's always been this problem of how do you communicate what you see in your head to your cast and crew. There are some things you have to see and feel. […] Now with this tool, I can share what I'm visualizing more clearly and efficiently to my creative team—the production designer, art designer, and cinematographer—for them to build on to enrich cinematic intelligence." We sat down with Martin Scorsese for a working storyboarding session with FLUX. More on Scorsese + BFL → https://bfl.ai/martin-scorsese-bfl-advisor Black Forest Labs is the frontier research lab building visual intelligence. Models that understand, reason, and act in the world.
FLUX.2: Frontier Visual Intelligence
0:53
Tutorial0:53
FLUX.2: Frontier Visual Intelligence
FLUX.2 is designed for real-world creative workflows, not just demos or party tricks. It generates high-quality images while maintaining character and style consistency across multiple reference images, following structured prompts, reading and writing complex text, adhering to brand guidelines, and reliably handling lighting, layouts, and logos. FLUX.2 can edit images at up to 4 megapixels while preserving detail and coherence. Learn more: https://bfl.ai/blog/flux-2
How to use the FLUX MCP.
6:02
Tutorial6:02
How to use the FLUX MCP.
The official FLUX MCP. Create images inside Claude, Cursor, Hermes, and more. Your agent will automatically route your request to the right model in the FLUX family and know exactly how to prompt it. From Klein for sub-second drafts to Pro for production-quality assets. Use it for website visuals, ad creatives, social content, product renders, concept art, or UI mockups. Watch Stephen’s short walk through of MCP setup and how he’s using it.

Who uses it

Real use cases, no hype
Text-to-Image Generation
Generate stunning, photorealistic images from text prompts.
Image Editing & Inpainting
Modify specific areas of an image using natural language.
Virtual Try-On
VTO tool for e-commerce catalog-scale garment try-on.
Multi-Reference Photography
Combine style, pose, and subject from up to 10 reference images.
Outpainting & Composition
Extend images beyond their borders with coherent scene generation.
Real-Time Apps (Klein)
Low-latency model deployment for interactive and real-time visual agents.
Synthetic Data Generation
Create high-quality annotated datasets for AI model training.
What's great
  • Top-tier photorealism and detail
  • Excellent prompt following
  • Multi-reference image editing
  • Apache 2.0 license for small models
Technical strengths
  • Flow-matching architecture
  • Dual CLIP/T5 encoding
  • Highly optimized for low-step generation
Where it falls short
  • High VRAM footprint on large models
  • Expensive for high-resolution batches
  • Worse complex prompt accuracy than competitors
  • No native BFL ControlNet
Technical limitations
  • T5-XXL encoder inference cost
  • Scheduler not SD-compatible
  • Distilled models lose fine-grained control
For technical folks

The deep-dive specs

Model Architecture
Transformer-based Diffusion
Max Resolution
4 Megapixels
Parameter Sizes
4B, 9B
Pricing Model
Pay-as-you-go per megapixel
API Support
REST API

FAQ

Includes technical Q&A
FLUX.2 introduces multi-reference image editing, better prompt adherence, and a newer architecture (Flow-matching 2.0), making it the recommended family over FLUX.1.

Traffic Insights

Monthly Visits

—

July 2026

Growth Rate

0%

vs last month

Dominance

0%

in Image Generation

Top Country

—

Top Countries

—

Growth Trend

Stable

Traffic has remained stable.

Reviews

What the internet actually thinks

Data refreshed weekly. No paid placements.

Trustpilot
Not reviewed yet
G2
G2
Not reviewed yet
C
Capterra
Not reviewed yet
Product Hunt
Not reviewed yet
COMMUNITY COMMENTS

Leave a Comment

Help others make informed decisions. Your honest feedback shapes the community's understanding of this tool.

Tips for commenting

  • •Be specific about features you used
  • •Share real use cases and results
  • •Mention both pros and cons
  • •Keep it honest and constructive
Sign in to leave a comment

Your comment will be published after moderation

Filter by rating:
Sort by:

No comments yet

Be the first to share your experience with this tool.

Alternatives

Not quite the right fit?

Here's what else is out there.

Fooocus
—
Unknown

Fooocus

Image Generation

Focus on prompts and images – no manual tweaking needed.

#ImageGeneration#StableDiffusion
View
Recraft
—
Freemium

Recraft

Image Generation

Create stunning visuals with AI – images, vectors, and mockups in seconds.

#AI#ImageGeneration
View
DALL·E 3
—
Freemium

DALL·E 3

Image Generation

Advanced text-to-image generation model that better understands and adheres to your prompts.

#ImageGeneration#AI
View
Midjourney
—
Paid

Midjourney

Image Generation

Generate stunning, photorealistic images and art with AI.

#AIGeneration#ImageGenerator
View
Dezgo
—
Freemium

Dezgo

Image Generation

Generate images and videos from text using the latest AI models, via web app or API.

#AIGeneration#StableDiffusion
View
Detail tags
Found via these searches
#AIGeneratedArt#DiffusionModels#APIIntegration#ImageEditing#OpenWeights#ComputerVision#CreativeAI