Downloads · 30 days
0
krystian-kaczor/krystian-flux-avatar
krystian-flux-avatar is a machine learning model from krystian-kaczor. Use it for the machine learning task on the model card, and read the license before you ship it in a product.
bash pip install diffusers transformers accelerate torch python import torch from diffusers import FluxPipeline
Downloads · 30 days
0
Access
Public
Updated Aug 7, 2025
Repo size
172 MB
Likes
0
Public
Click a slice to open those files.
.safetensors172 MB · 100%
From the Hugging Face model README
---
license: creativeml-openrail-m
base_model: black-forest-labs/FLUX.1-schnell
tags:
- flux
- lora
- text-to-image
- diffusers
- avatar
instance_prompt: "a photo of KAKI"
---
# LoRA for FLUX: krystian-flux-avatar
These are LoRA weights for the base model **[black-forest-labs/FLUX.1-schnell](https://huggingface.co/black-forest-labs/FLUX.1-schnell)**. The weights were trained on a custom dataset to generate images in a specific style, triggered by the word `KAKI`.
## How to use
First, make sure you have the `diffusers` library installed:
```bash
pip install diffusers transformers accelerate torch
Then, you can use the following Python code to load the base model and apply the LoRA weights.
import torch
from diffusers import FluxPipeline
# Load the base FLUX model
model_id = "black-forest-labs/FLUX.1-schnell"
pipe = FluxPipeline.from_pretrained(model_id, torch_dtype=torch.bfloat16)
pipe.to("cuda")
# Load the LoRA weights into the pipeline
lora_repo_id = "krystian-kaczor/krystian-flux-avatar"
pipe.load_lora_weights(lora_repo_id)
# Generate an image using the trigger word
prompt = "photo of a beautiful woman, KAKI style, cinematic, detailed, high resolution"
# Generate the image
image = pipe(
prompt=prompt,
num_inference_steps=8,
guidance_scale=0,
).images[0]
# Save the image
image.save("generated_image.png")
print("Image saved as generated_image.png")
You should use KAKI in your prompt to trigger the LoRA. For best results, you can use it as "KAKI style" or simply "KAKI".
These weights were trained on Replicate using the following parameters:
krystian-kaczor/train-flux-lora1000KAKI999krystian-kaczor/krystian-flux-avatar