Instructions to use lvladikov/Krea2-Turbo-Distill-4step-LoRA with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Diffusers
How to use lvladikov/Krea2-Turbo-Distill-4step-LoRA with Diffusers:
pip install -U diffusers transformers accelerate
import torch from diffusers import DiffusionPipeline # switch to "mps" for apple devices pipe = DiffusionPipeline.from_pretrained("krea/Krea-2-Turbo", dtype=torch.bfloat16, device_map="cuda") pipe.load_lora_weights("lvladikov/Krea2-Turbo-Distill-4step-LoRA") prompt = "Astronaut in a jungle, cold color palette, muted colors, detailed, 8k" image = pipe(prompt).images[0] - Inference
- Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- Draw Things
- DiffusionBee
Krea 2 Turbo β 4-Step Distillation LoRA
π New: a 3-step LoRA takes Krea 2 Turbo from 8 steps down to 3, at three quarters of this LoRA's denoising time, with more fine detail (fine texture 1.26Γ the 8-step teacher's on its 22 test prompts). This 4-step LoRA stays the closest to Turbo's own 8-step pictures.
Half the steps Β· ~1.6Γ faster end to end Β· cleaner than the first release: grid patterns, ghosting and flat-area grain all closer to the 8-step teacher Β· 13 trained resolutions up to 1824Γ1024 Β· 133,000 training samples, 1.7Γ the first release.
π This version (chk00133000) replaces the first 'final' release (chk00078000), from two months ago. I decided to train a 'few more' samples (55K more to be precise), the newer part with five critic heads where the first release had one. Measured against the 8-step teacher on the 15 test prompts at 12 sizes:
- cleaner fine structure β the 8- and 16-pixel grid patterns at 0.96β0.98Γ the teacher's, where the first release was at 1.09β1.11Γ
- less ghosting and fewer doubled edges β closer to the teacher at 10 of 12 sizes on both measures
- cleaner flat areas β grain at 0.99Γ the teacher's, down from 1.08Γ
- no over-sharpening at large sizes β fine texture at 1.01β1.04Γ the teacher's at 1440 px, down from 1.18β1.20Γ
- saturation matching the teacher's
Fine texture now sits at the teacher's level, a touch softer than the first release at mid sizes. The first release stays in
_archive/checkpoints/.
A LoRA for Krea 2 Turbo that reduces the minimum usable step count from 8 to 4 β Turbo's own model and sigmas, half the denoising passes, and the fine texture that 4-step Turbo loses put back.
- β‘ Half the steps β 8 β 4, on Turbo's own deployment sigmas
- β±οΈ ~1.6Γ faster end to end β 54.5 s vs 88.7 s at 1024Γ1024 for the whole run, model load, text encoding and VAE decode included (1.8Γ on denoise alone, without them)
- π― Fine texture at the teacher's level β 0.96Γ the teacher's fine-texture energy on average over 12 sizes, 1.01β1.04Γ at 1440 px; the 8- and 16-pixel grid bands 0.96β0.98Γ, flat-area grain 0.99Γ, saturation 1.00Γ
- π£οΈ Prompt-aware training β critic scores images against their prompts during training
- π 13 trained resolutions β multi-aspect from 512Γ512 up to 1440Γ1440, and 1824Γ1024 since the first release
- π Drop-in β plain LoRA weights for diffusers and ComfyUI. No custom nodes, no patched sampler, no code
- π² 13,750 prompts from Lakonik's 3M-prompt dataset, each recorded as a full 8-step teacher trajectory; since the first release, 91,044 training chords on Turbo's own 8-step paths
- π· 43,044 real-photo crops β 25,560 from LSDIR + 17,484 from Flickr2K, captioned, in critic's real set; since the first release, 5,234 crops cut from 4K photos
- π’ 133,000 training samples in shipped weights β 1.7Γ the first release's 78,000
- π 21 days on a single RTX 3090 to the first release; since then about a day on an H200 and two more on the RTX 3090
- π 16 recipe adjustments β each made on measurement of the one before; since the first release, five critic heads instead of one
π‘ Too strong on a prompt? Turn it down. The LoRA restores fine texture, and on some subjects β stylised art, high-contrast splash pieces, very large renders β that can read as too much at strength 1.0. Effect scales smoothly: 0.75 is a good second setting, 0.6β0.9 is fair game.
Files
| file | what it is |
|---|---|
krea2_turbo_4step_rank_64_lora.safetensors |
LoRA in diffusers key format β see diffusers |
krea2_turbo_4step_rank_64_lora_comfyui.safetensors |
Same weights under ComfyUI key names β see ComfyUI |
krea2_turbo_4step_lora_t2i.json |
Ready ComfyUI workflow, stock nodes only |
krea2_turbo_4step_rank_64_lora_checkpoint_info.md |
Which checkpoint the two weight files are, with their SHA-256s |
LICENSE.pdf |
Krea 2 Community License Agreement |
NOTICE.txt |
Required attribution notice |
Both weight files are one adapter β only key names differ. Both carry training details in safetensors metadata.
Quick Start
diffusers
pip install git+https://github.com/huggingface/diffusers.git
import torch
from diffusers import Krea2Pipeline
from huggingface_hub import hf_hub_download
from safetensors.torch import load_file
pipe = Krea2Pipeline.from_pretrained("krea/Krea-2-Turbo", torch_dtype=torch.bfloat16).to("cuda")
lora = hf_hub_download("lvladikov/Krea2-Turbo-Distill-4step-LoRA", "krea2_turbo_4step_rank_64_lora.safetensors")
state = load_file(lora)
state = {f"transformer.{k}": v for k, v in state.items() if not k.endswith(".alpha")}
pipe.load_lora_weights(state, adapter_name="4step")
image = pipe("a fox in the snow", num_inference_steps=4, guidance_scale=0.0).images[0]
image.save("krea2_4step.png")
num_inference_steps=4is the whole config. Pipeline applies Turbo's fixed timestep shift (ΞΌ = 1.15) and evaluates at Ο = 1.0, 0.905, 0.760, 0.513 β exactly the four points the LoRA was trained on. Keepguidance_scale=0.0.- Strength:
pipe.set_adapters(["4step"], adapter_weights=[0.75]). Stock Turbo for comparison:pipe.unload_lora_weights()+num_inference_steps=8.
ComfyUI
| file | put it in |
|---|---|
krea2_turbo_4step_rank_64_lora_comfyui.safetensors |
ComfyUI/models/loras/ |
krea2_turbo_bf16.safetensors (from Comfy-Org/Krea-2) |
ComfyUI/models/diffusion_models/ |
qwen3vl_4b_bf16.safetensors (same repo) |
ComfyUI/models/text_encoders/ |
qwen_image_vae.safetensors (same repo) |
ComfyUI/models/vae/ |
Load krea2_turbo_4step_lora_t2i.json. Full bf16, no quantisation, runs on CUDA / Apple Silicon / CPU unchanged.
Settings: steps 4, cfg 1.0, sampler euler / simple, LoRA strength 1.0.
βοΈ cfg 1.0, not 0.0. ComfyUI expresses "no CFG" as 1.0 (one forward pass); diffusers uses 0.0. Setting 0.0 in ComfyUI is not the same thing.
LoRA Strength
| strength | what happens |
|---|---|
| below 1.0 | Correction partly applied β output between unassisted 4-step and full LoRA. 0.75 for when 1.0 is too textured/hard |
| 1.0 | Trained point, recommended |
| 1.0β1.5 | Extrapolation β surface detail denser, micro-contrast harder, coherent but stylised |
| above 1.5 | Not recommended β breaks into uniform speckle/noise over whole image |
Reach for steps before strength: 1.0 at more steps is the dependable way to get more. Strength and step count trade against each other.
Hardware
Trained on single RTX 3090 (24 GB) to the first release; since then about a day on a single H200 and two more on the RTX 3090. Frozen base quantized weight-only to int8 (blockwise-64) β ~0.007 relative error, ~3% step time cost. Teacher trajectories rolled at same precision (targets permanent, error baked in).
Every bucket trained under full int8 up to 1440Γ1440. Large buckets fit via:
- Checkpointed block inputs staged to pinned host memory (numerically exact)
- Discriminator pass after generator backward (peaks don't overlap)
- Trunk stopped at feature tap it actually reads
Released LoRA is bf16 applied to unquantized base.
Usage Notes
- π― Krea 2 Turbo only β trained against Turbo's weights and schedule
- π« Keep guidance at 0.0 (ComfyUI: cfg 1.0, not 0.0)
- π Keep ΞΌ = 1.15 β training targets anchored to that grid
- π¬ Training used int8 base; released LoRA is bf16 on unquantized base
Examples
Every sheet below: base model (8 steps), base at 4 steps without LoRA, base at 4 steps with LoRA β same seed throughout. Compare panels 2 vs 3 to isolate LoRA effect. NFE = steps (Turbo is CFG-free).
Portrait of a young woman with freckles and windswept auburn hair...
Kingfisher bursting out of water...
Rainy night city street with glowing neon signs...
(12 more examples in full version β see assets/ for all 15 prompts)
Resolution Sweeps
assets/resolution_sweeps/ β this LoRA at every trained resolution for all 15 test prompts (same prompts, seed, 4 steps, strength 1.0), 1824Γ1024 included. Nothing cherry-picked. This version's sweep was rendered in ComfyUI (seed 4242); the teacher's renders came from another renderer, so pose and framing can differ between the two.
_teacher-8step/ β official Krea 2 Turbo 8-step reference renders for same prompts/seeds/resolutions.
Layout:
assets/resolution_sweeps/
βββ _teacher-8step/ 8-step stock Turbo reference
β βββ 512x512/
β βββ 1024x1024/
β βββ ... (all 12 buckets)
βββ 4step-LoRA/ Same tree, this LoRA at 4 steps (+ 1824x1024)
Two ways to read: down the sweep (does it hold across resolutions?) or against teacher (open same file in both folders).
Archive
Every published checkpoint and its resolution sweep under _archive/ β checkpoints/ and resolution_sweeps/, each under its number: chk00133000, this release, the first 'final' release chk00078000 (its sweep in chk78000-final/), and the earlier chk00060000, chk00042000, chk00026000, chk00019000, chk00014000, chk00010000, chk00006000, chk00005000 and chk00004000.
Detailed Model Card
The full write-up is in the Detailed Model Card:
- How I got here β from stock Turbo to the first release, and its timeline
- chk00133000 vs chk00078000 β what this release changed against the first one
- Measured against the teacher β the first release's measurements: texture, the 4-step gap, the judged renders
- This is not a RawβTurbo diff β and using it on Raw
- Inference with diffusers and ComfyUI β the settings explained
- Performance β the time per model call and per render
- LoRA strength β where the useful range is
- File format and compatibility β the keys, runtime by runtime
- Resolution sweeps β every trained size, every test prompt
- Method β the progressive-distillation targets and the critic
- What the LoRA touches β the 228 modules
- Training data β the recorded trajectories and the photo crops
- Resolutions β the share of training at each size
- Hardware
- Examples β all 15 prompts against the teacher, and why these prompts
- What's next β the 2-step and 3-step LoRAs that followed
License
This adapter is a Derivative of Krea 2 Turbo under the Krea 2 Community License Agreement. Everything the agreement says about Krea 2 Turbo applies to this LoRA: Acceptable Use Policy, revenue threshold for commercial use, content-filtering duty for deployments.
Copy of agreement: LICENSE.pdf, required notice: NOTICE.txt. See krea.ai/krea-2-licensing.
Not an official Krea product, not endorsed by Krea. Base model is Krea's; adapter weights and everything else in this repo are mine.
- Downloads last month
- 18,656