This Lora is extraxted from a fine tuned checkpoint, based on Flux dev de-distilled.
This checkpoint has been trained on high resolution images that have been processed to enable the fine-tune to train on every single detail of the original image, thus working around the 1024x1204 limitation, enabling the model to produce very fine details during tiled upscales that can hold up even in 32K upscales. The result, extremely detailed and realistic skin and overall realism at an unprecedented scale.
This first alpha version has been trained on male subjects only but elements like skin details will likely partically carry over though not confirmed.
Training for female subjects happening as we speak.
CFG: 2-4
Steps: 50-60
With Turbo and Fast Lora
CFG: 2-4
Steps: 8-16
Turbo Lora weight: 1
Fast Lora weight: 0.33
🧪 Example Usage
Use the following Python code to generate images with this LoRA using the Diffusers library:
python
1import torch
2from diffusers import FluxPipeline
34# Load the base model5pipe = FluxPipeline.from_pretrained(6"black-forest-labs/FLUX.1-schnell",7 torch_dtype=torch.float16,8 use_auth_token=True# Make sure you're logged into Hugging Face CLI9).to("cuda")1011# Load the LoRA weights from Hugging Face Hub12pipe.load_lora_weights("shemayons/Flux-Sigma-Vision-Lora", weight_name="flux_sigma_vision_lora.safetensors")1314# Optionally fuse LoRA weights with a custom scale15pipe.fuse_lora(lora_scale=0.8)1617# Define your prompt18prompt ="a futuristic cityscape at sunset"19negative_prompt ="low quality, blurry"2021# Generate the image22image = pipe(prompt=prompt, negative_prompt=negative_prompt, num_inference_steps=30).images[0]2324# Show the result25image.show()26
This Flux variant works with negative prompts so use them if you want to. Negative prompt will work with and without the Turbo Lora.
Download model
Weights for this model are available in Safetensors format.
