FLUX.1 Kontext Dev
lastmod 2025-11-27created 2025-06-294 edits
Open weights for image editing
About
- FLUX.1 Kontext [dev] released for free as an open-weight version of a previously proprietary generative image editing model
- With 12B parameters it can run on consumer-grade hardware, and is distributed for free research and non-commercial use
- Offers commercial-grade, high-quality image editing such as fine-grained local/global edits, character consistency, and iterative refinement, and integrates out of the box with popular frameworks such as ComfyUI, Hugging Face Diffusers, and TensorRT
- In evaluations, outperforms existing open and closed models (e.g., Google's Gemini-Flash Image)
- Also provides TensorRT variants optimized for the NVIDIA Blackwell architecture (BF16, FP8, FP4), delivering much higher speed and efficiency on the latest hardware
- Businesses can also easily integrate and use it through a commercial license and a self-serve portal
Resources
- Model weights: https://huggingface.co/black-forest-labs/FLUX.1-Kontext-dev
- Code: https://github.com/black-forest-labs/flux
- API Documentation: https://docs.bfl.ai/quick_start/introduction
- Self-Serve Portal: http://bfl.ai/pricing/licensing
- Helpdesk: https://help.bfl.ai