Using Flux Kontext in Krita with the Generative AI Plugin

Channel:
Subscribers:
2,970
Published on ● Video Link: https://www.youtube.com/watch?v=sirXwDmVmEM



Duration: 0:00
5,077 views
0


A look at using the Flux Kontext model within Krita (free/open source) with the AI diffusion plugin for instruction-driven image-to-image. Flux Kontext is a new sort of image generation model - you tell it want you want it to do, not what you want in the final image.

Kontext understands the difference between "keep this" and "change that," making it incredibly impressive for maintaining character consistency across multiple edits. This is mostly an overview of the setup, some demo images, working around some quirks in Krita, futzing with the Krita's layers, and trying to show some additional capabilities of Kontext. Is this a workable Stable Diffusion XL alternative? Absolutely. Can this replace Adobe Photoshop's generative AI tools? In many cases, probably.

edit: update -- these quants work out of the box with Krita as well. I just tried Q4_0. It's.. alright. I think the fp8 one linked on the GitHub page works better. Q4_0 seems to follow the prompt worse. The images still look great, but they're less close to what was asked for than the output from the larger models. You'll still need the text encoder and VAE when using the GGUF quants.

Krita
https://krita.org/en/

Krita AI Diffusion Plugin by Acly
https://github.com/Acly/krita-ai-diffusion

Click view PDF for a link to the Flux Kontext evaluation paper
https://arxiv.org/abs/2506.15742

Flux Kontext Prompting Guide
https://docs.bfl.ai/guides/prompting_guide_kontext_i2i

Flux Kontext models compatible with the AI diffusion plugin (out of the box)
https://huggingface.co/Comfy-Org/Lumina_Image_2.0_Repackaged/blob/main/split_files/vae/ae.safetensors
https://huggingface.co/comfyanonymous/flux_text_encoders/blob/main/clip_l.safetensors
https://huggingface.co/comfyanonymous/flux_text_encoders/resolve/main/t5xxl_fp8_e4m3fn_scaled.safetensors

[0:00] What Even Is This And Why Are We Here?
[1:10] Model Capabilities and Tasks
[2:15] Visual Cues and Masking
[2:42] Installation Tips
[4:49] Example Flux Kontext Workflow in Krita
[7:23] Exploring Flux Kontext Edge Cases
[8:50] High Resolution Compositing with Flux Kontext




Other Videos By NanoNomad


2025-08-09Saturday Morning Console Wars: 40 Minutes of Restored Retro Console Commercials
2025-08-06MS-DOS and Windows XP Gaming on a Thinkpad X61 [SoundBlaster Emulation with MIDI in DOS]
2025-07-11Using Flux Kontext in Krita with the Generative AI Plugin
2025-04-17DiffRhythm: Generative Music (done quickly)
2025-02-25Is YuE the Stable Diffusion of Music? | Generate Full-Length Songs with Vocals at Home
2025-02-10Portable Whisper Speech to Text with Speaker Diarization and VAD | Purfview Faster Whisper XXL
2024-07-03Fine Tuning XTTS v2 for Hindi Speech with forked Coqui TTS
2024-06-26Fine Tuning XTTS v2 with forked Coqui | Coqui AI is dead; Long live Coqui!
2024-06-202x Faster LLM Training on Windows | LLaMA-Factory with Unsloth and Flash Attention 2
2024-06-1564kb Scene Demo/Intro/Cracktro Multimedia Mix #1 (90 min) | Flash/Photo-sensitivity Warning
2024-06-10Stable Audio Open 1.0 | Open Source* Generative Audio and Fine Tuning*
2024-06-04Troubleshooting Sega Saturn Emulation with Retroarch for iOS/Apple
2024-05-29Play Windows 98 and MS-DOS Games on iPad/iOS/iPhone with DOSBox-Pure and Retroarch for FREE
2024-05-25The Lost Art of Optical Disc Repair | Fixing and Testing a PlayStation Disc
2024-05-22Retroarch iOS Updates | Improved Performance, MS-DOS Core, Doom and Touch Input
2024-05-17RetroArch for iPad and iPhone now on the App Store | Installation, Setup, Quick Performance Overview
2024-05-13Micca Speck 4K Media Player | Unboxing, Firmware Update, Setup, Demos, and Opinions
2024-05-06Training SDXL to Generate Text Using IA3 LoRA | It's like Kai's Power Tools, I Guess?
2024-04-17Replacing Faulty Asus Phoenix RTX 3060 GPU Cooler - It's Easy
2024-03-21Bark TTS, Seamless Translation, RVC, Music Generation and More with the TTS Generation WebUI
2024-02-14Train Better Stable Diffusion Models | Prep Datasets Using this Free "Magic" Image Tool