NOW available On-Site! The LoRA version is NOT a checkpoint, but a LoRA for local gen.

μKrea2 (mu - Krea2) is my own final Krea2 mix based on several trained LoRAs (including nKrea) and some extracted LoRAs made as an experiment of hacking sd_mecha and learning how to unquantize INT8-ConvRot checkpoints (because undoing stuff is the best way to learn of to make stuff).
And since both the following events happened:
I finally found out how to convert stuff to INT8-ConvRot (via convert-to-quant),
stable-diffusion.cpp is now compatible with INT8-ConvRot,
here it is for your enjoyment!
I am releasing it as Early Access at low Buzz price to avoid immediate scraping.
Since this is Krea2 checkpoint, nothing special on usage, here is just some info on how i use it (for the showcase pictures for example):
ER_SDE/Smoothstep (aka Beta)
CFG 1 - 8 Steps
VAE: HDR VAE by @Felldude (basically the best FP32 VAE for WAN2.1/ANIMA/Krea2/Qwen Image)
All pictures made using stable-diffusion.cpp version master-874-656a135-2-g2ea8aff, commit 2ea8aff, compiled from source on ubuntu 26.04 with CUDA 13.2 with an RTX 3060 - 12GB. Each 960x1440 px pictures generated in around 60s (including decode).
Command line used (to start server):
./sd-server -l 0.0.0.0 -p "" -n " " --sampling-method er_sde --scheduler smoothstep --steps 8 -W 960 -H 1440 -b 1 --cfg-scale 1 -s -1 --clip-skip -1 -t 0 --rng cuda --sampler-rng cuda --diffusion-model muKrea2-INT8-convrot.safetensors --vae finalcut_vae.safetensors --llm Qwen3VL-4B-Instruct-Q8_0.gguf --vae-tile-overlap 0.5 --vae-tile-size 32x32 --offload-to-cpu --vae-tiling --fa --eager-loadAll pictures batched using a small custom sd-client that talk to the server using the API endpoint (i'll release it if someone wants it).
./sd-client --help
Usage: sd-client [-s "832x1216"] -o output.png -p "A picture of a cat"Thanks for reading!
Special thanks (for the great pictures they make that helped me find a direction for this):
@llikswonskcalb
@ElectricDreams
@Chumple
@lonecatone23
@Sophorium
@daceheg192491
@Liant
PS, picture in the model card prompt: "A female satyr, naked human woman upper body and face, wearing a latex black bikini. goat lower body, with furry legs and goat hooves, horns and ears. Bright orange curly hair, wide smile, horizontal pupils. Lying down on her back in the grass, seen from above. A broken sign saying "WARNING, UNPROMPTED NSFW CAN HAPPEN" is visible on the ground next to her. Warm sunlight, cheerful vibe. Safe."
Description
Posted on Civitai, upload found elsewhere are FORBIDDEN and done without the author permission.
FAQ
Comments (15)
New toy right in time for my latest workflow round 👍
💜
Great checkpoint, very good images!
Thank you! 💜
🛡️ Passed the official BSS (Bullshit Screening Service) inspection with flying colors. Zero slop detected. Must-download! :3
Thank you! <3
DesireX is also such a great model, the showcase pictures are perfect <3
wow, testing now very nice results, :)
Thank you!
Quote perfect, just a bit wet
Oh yes, that it is 😂
I may have gone overboard when doing my training 🤔😅
argh.. told myself to stop collecting models for sake of my SSD.. But couldn't resist. thanks for the great model.
I can always propose a LoRA version on top, this could save some space, Krea2 is rather... invasive in term of storage 😭
There is now a "LORA VERSION" which a 128 dim LoRA extract of this model against Krea2 Turbo. It will not be "the same" but still, ~900MB vs 13GB, that could help saving your SSD :D
Amazing quality, thanks for sharing!
With Early access ending, the model turned to a 100 Buzz paying price for a few hours. Sorry, it was not my intention, it was done automatically. It is now free for everyone :D



















