FLUX.1 Kontext NVIDIA NIM Microservice Available Now

RTX AI

Black Forest Labs’ FLUX.1 Kontext [dev] image editing model is now available as an NVIDIA NIM microservice, bringing simple, language-based image editing to RTX AI PCs without complex workflows or fine-tuning.

This release streamlines the often-complicated process of deploying AI, from curating model variants and adapting data pipelines to quantization for reduced VRAM use. The result is faster, more efficient generative AI workflows for creators, developers, and enthusiasts alike.

More Intelligent Image Editing with Generative AI

FLUX.1 Kontext [dev] is an open-weight generative model built for precision image editing. Its guided, step-by-step generation process gives users fine-grained control from tweaking small details to reimagining entire scenes.

Generative AI

By accepting both text and image inputs, the model enables natural, intuitive edits that preserve the core visual concept. Simple language prompts replace complex coding, making professional-quality edits accessible to more users.

Model

Optimized for Speed and RTX AI PCs

Thanks to the collaboration between NVIDIA and Black Forest Labs, FLUX.1 Kontext [dev] has been quantized for smaller model sizes and faster performance.

  • FP8 version: 12GB, optimized for GeForce RTX 40 Series GPUs
  • FP4 version: 7GB, optimized for GeForce RTX 50 Series GPUs using SVDQuant for quality preservation

When paired with NVIDIA TensorRT, the model achieves over 2x acceleration versus the original BF16 version in PyTorch, delivering performance gains once reserved for AI specialists.

PyTorch

Getting Started with FLUX.1 Kontext [dev]

FLUX.1 Kontext [dev] is available on Hugging Face with TensorRT optimizations and integrates seamlessly with ComfyUI NIM nodes.

Kontext

Quick setup steps.

  1. Install NVIDIA AI Workbench
  2. Get ComfyUI
  3. Install NIM nodes via ComfyUI Manager
  4. Accept the model license on Hugging Face
  5. Click Run to start editing

NIM microservices are optimized for NVIDIA RTX GPUs and feature models from across the AI community. Explore more on GitHub and build.nvidia.com, and join the RTX AI community on Discord, Facebook, Instagram, TikTok, and X for the latest AI tools and workflows.