Back to products
FLUX.1 Kontext

FLUX.1 Kontext

Powerful In-context AI image editing, now open source

Overview

What it is

FLUX.1 Kontext by Black Forest Labs is a powerful model for in-context image editing. The 12B [dev] version is now open-weight, offering top-tier character and style consistency on consumer hardware.

Intent

I need it when

Manipulate text within images and ensure typography blends naturally with surrounding visual content

FLUX.1 Kontext includes typography manipulation capabilities, allowing users to change text within images while ensuring seamless integration with the surrounding scene. This is useful for design work, localization, and creative projects requiring text modifications.

Make targeted edits to specific regions of an image without affecting surrounding content

FLUX.1 Kontext supports local editing, allowing users to make precise modifications to specific image elements through text instructions. This enables selective changes (e.g., modifying one object) while preserving the rest of the image composition and quality.

Edit images by describing changes in natural language while preserving character identity and scene context

FLUX.1 Kontext performs in-context image generation and editing by accepting both text prompts and reference images. Users can iteratively modify images (remove objects, change environments, adjust weather) while maintaining character consistency and contextual relationships, enabling fast creative workflows without complex manual editing.

Deploy image generation and editing on consumer hardware without cloud dependency

FLUX.1 Kontext [dev] is an open-weights 12B parameter model that runs on consumer hardware, providing proprietary-level image editing performance without requiring API access. Users can download code from GitHub and self-host for full control and privacy.

Generate new scenes featuring the same character or object across multiple environments while maintaining visual consistency

FLUX.1 Kontext's character consistency feature preserves unique identity elements across multiple scenes and environments. Users provide a reference character or object and text prompts describing new scenarios, and the model generates coherent variations that maintain the subject's visual identity.

Drop

Not a fit when

  • User needs fixed monthly subscription pricing with predictable costs; FLUX.1 Kontext uses pay-as-you-go per-image pricing
  • User requires offline-only image editing without API dependency; FLUX.1 Kontext [pro] and [max] require API access (only [dev] can run locally)
  • User needs traditional text-to-image generation without image input capability; FLUX.1 Kontext is optimized for in-context editing combining text and images
  • User requires guaranteed SLA and dedicated support for small projects; enterprise features require contacting sales
  • User needs real-time image editing with sub-second latency; FLUX.1 Kontext prioritizes quality and consistency over extreme speed
Commercials

Pricing

Pay-as-you-go per image generated. Three model tiers available: FLUX.1 Kontext [dev] (open-weights, 12B parameter), FLUX.1 Kontext [pro] (fast iterative editing), and FLUX.1 Kontext [max] (premium performance). Enterprise custom pricing available. View pricing