Update:
- V3 is about half RAW still runs at 8-12 steps.
- Added V3 FP8.
- Added V3 Int4 ConvRot SR (stochastic rounding).
⚠️ IMPORTANT — Applies to v1.0, v1.1,v2, v3 ⚠️
❌ Do NOT use the INT8 Fast custom node.
This checkpoint is already in native ComfyUI INT8 (ConvRot) format.
✅ Use the standard ComfyUI Load Diffusion Model node.
Requirements
- ComfyUI v0.27+
- ComfyUI Kitchen
Krea2 Center v1: similar, experimenting with some block merging.
Krea2 Center v0.5 explores whether Krea2 can become more expressive through carefully balanced LoRA merging while preserving the model's realism, flexibility, and overall character. The goal isn't to replace Krea, but to gently extend its emotional range across portraits, fantasy, sci-fi, and stylized scenes.
Added Int8: about 25% faster on 50 series Nvidia don't have a 40 or 30 series to test.
Krea 2 is licensed under the Krea 2 Community License Agreement. For more information, visit https://krea.ai/krea-2-licensing.
Description
int4 convrot, requires latest Comfyui pull
FAQ
Comments (8)
Nice, looks like I should be able to run INT4 in my 3060, will have to try it.
When I use the UNet loader, I get an [ERROR] Got an OOM, unloading all loaded models. How do you run this model?
are you on the latest Comfyui commit? I suppose you are talking about the int4?
@TiwazM comfyui 0.27， int8 2.0
@woaiyitiaofeng with comfy kitchen installed?
@TiwazM installed, not sure why. I'm using a 3080, so it should be fine. You can try this version
@woaiyitiaofeng not sure I have a way to test it it runs fine for me very latest comfy ui. I dont have a 30 series card to confirm. I did try the v2 int8 in forge classic neo works fine there too.
@woaiyitiaofeng I will upload an fp8 probably today just in case.