The Flux Series is a comprehensive family of AI image generation models. Across the lineup, Flux supports text-to-image, image-to-image, reconstruction, contextual reasoning, and high-speed creative workflows.

Key Features

High-quality image generation
Produces detailed, visually appealing images suitable for both creative exploration and professional production.
Fast inference options
Ultra-Fast and Schnell variants deliver rapid generation, ideal for interactive tools and iterative design.
Strong contextual understanding
Kontext models interpret prompts and multi-input relationships more accurately, improving prompt-following and composition.
Reliable output stability
Designed for consistent structure, style, and subject preservation across generations.

Flux 2 Series (Next-Generation High-Fidelity Models)

FLUX 2 PRO
FLUX 2 DEV
FLUX 2 FLEX
FLUX 2 DEV LoRA
FLUX 2 MAX
FLUX 2 FLASH
FLUX 2 TURBO
FLUX 2 KLEIN

Flux Dev Series (Core Foundation Models)

flux-dev
flux-dev-ultra-fast
flux-dev-lora
flux-dev-lora-ultra-fast
flux-dev-lora-trainer
flux-dev-lora-trainer-turbo

Why Choose the Flux Dev Series?

Strong general-purpose generation capabilities
Fast, LoRA-ready, and training-focused variants
Stable and production-friendly foundations

Flux Redux & Fill Series (Editing & Reconstruction)

flux-redux-dev
flux-redux-pro
flux-fill-dev

Why Choose Redux & Fill?

Tailored for image editing, variation, and reconstruction
Supports guided restyling and localized adjustments

Flux Kontext Series (Contextual & Multi-Input Models)

Why Choose the Kontext Series?

Enhanced contextual comprehension
Multi-input support for complex workflows
Available in Dev, Pro, and Max performance tiers
Ultra-Fast versions for low-latency usage

Flux 1.1 Pro Series (High-Quality Generation)

flux-1.1-pro
flux-1.1-pro-ultra

Why Choose Flux 1.1 Pro?

Provides consistently clean, high-quality images
Suitable for commercial content creation
Ultra versions balance speed and quality

Flux Schnell Series (Speed-Optimized Models)

flux-schnell
flux-schnell-lora

Why Choose Flux Schnell?

Designed for ultra-fast generation
Ideal for real-time tools and rapid prototyping
LoRA variants enable fast customization

Flux SRPO Series (Image Enhancement & Refinement)

flux-srpo
flux-srpo/image-to-image
flux-1-srpo
flux-1-srpo/image-to-image

Why Choose SRPO?

Suitable for enhancement and refinement
Improves clarity and structure in image-to-image tasks

Flux ControlNet Union Pro 2.0

flux-controlnet-union-pro-2.0

Use Cases

Controlled generation guided by structural inputs
Ideal for design and layout-specific workflows

Krea & Pulid Series

flux-krea-dev-lora
flux-pulid

Toy Me Image-To-Video Effect

Toy Me turns photos into toy-figurine videos with smooth morphs and detailed textures for playful image-to-video effects. Ready-to-use REST inference API, best performance, no coldstarts, affordable pricing.

ExemplesTout voir