Flux Kontext Pro
Overview
Flux Kontext Pro is an instruction-based image editing model developed by Black Forest Labs. It allows creators to modify existing images using natural language prompts rather than manual masking or inpainting. The model is especially good for professional workflows requiring precise, context-aware transformations—such as rebuilding environments or adding objects—while strictly preserving the original image's structure and photorealistic details. For users needing the highest possible performance, Black Forest Labs also offers Flux Kontext Max.
Flux Kontext Pro Image to Image
Image → Image (edit) generation.
Specifications
- Input mode
- Image → Image (edit)
- Aspect ratios
- 21:9, 16:9, 4:3, 3:2, 1:1, 2:3, 3:4, 9:16, 9:21
- Pricing
- 5 credits / image — higher resolutions cost more
- Typical generation time
- ~16s
- Free tier
- No
Image → Image (edit) examples
Flux Kontext Pro Text to Image
Text → Image generation.
Specifications
- Input mode
- Text → Image
- Aspect ratios
- 21:9, 16:9, 4:3, 3:2, 1:1, 2:3, 3:4, 9:16, 9:21
- Pricing
- 5 credits / image — higher resolutions cost more
- Typical generation time
- ~16s
- Free tier
- No
Text → Image examples
What is Flux Kontext Pro best used for?
Flux Kontext Pro is designed for instruction-based image editing and style transfer using natural language. Instead of relying on manual masking or traditional inpainting, users can tell the model what to change—such as adding an object, swapping an outfit, or altering a scene's weather. The model preserves the original image's character consistency and untouched details, making it useful for rapid prototyping and iterative visual adjustments.
Who developed Flux Kontext Pro and when was it released?
Flux Kontext Pro was developed by Black Forest Labs and announced in May 2025. It builds upon the text-to-image architecture of earlier models like Flux 1.1 Pro and Flux 1.1 Ultra by introducing native multimodal editing capabilities. It launched alongside an open-weights Dev version and the premium Flux Kontext Max, which provides higher fidelity for complex tasks like typography. The company later advanced this lineage with the [Flux.2 pro] series.
How can I get the best results when editing with Flux Kontext Pro?
Take advantage of the model's visual consistency by using multi-turn editing. Rather than packing complex changes into a single prompt, refine your image through successive, isolated instructions—for example, changing the background in one step and modifying clothing in the next. When integrating the model into automated workflows, explicitly configure your requests to match the input image's aspect ratio to prevent unwanted cropping. For more advanced implementations, refer to the official inference repository.
Similar models
Prompt tips
- Use conversational instructions: Write direct commands like "Change the orange flowers to a bouquet of roses" rather than traditional comma-separated keywords.
- Explicitly preserve elements: State what should remain untouched (e.g., "maintain the background" or "keep the facial features identical") to prevent the model from hallucinating unwanted changes.
- Choose verbs carefully: The model reacts differently to specific action words; "change," "replace," and "transform" will yield varying degrees of structural alteration.
- Iterate in steps: Break complex transformations into smaller, sequential edits rather than overloading a single prompt, taking advantage of the model's multi-turn capabilities.
