ec4m / README.md
aimalias's picture
Upload README.md with huggingface_hub
eb840d9 verified
|
Raw History Blame Contribute Delete
1.96 kB
---
base_model: krea/Krea-2-Raw
tags:
- text-to-image
- diffusers
- lora
- krea2
- template:sd-lora
license: apache-2.0
instance_prompt: "EC4M woman"
widget:
- text: "A cinematic shot of an EC4M woman wearing bioluminescent armor, standing amidst a floating neon jungle on a distant alien planet."
output:
url: sample_0.png
- text: "A nostalgic 1950s oil painting of an EC4M woman reading a leather-bound book in a cozy, sun-drenched library filled with dust motes."
output:
url: sample_1.png
- text: "A high-fashion editorial photo of an EC4M woman draped in iridescent silk, posing against the stark, brutalist concrete architecture of a futuristic city."
output:
url: sample_2.png
---
# Krea 2 LoRA — aimalias/ec4m
<Gallery />
A DreamBooth-LoRA for **Krea 2**, trained on **Krea 2 RAW** and shown on **Krea 2 Turbo**. The samples below were generated with this LoRA on Turbo (8 steps).
## Trigger
Use the token `EC4M woman` to invoke the concept.
## Samples
![sample](./sample_0.png)
> *"A cinematic shot of an EC4M woman wearing bioluminescent armor, standing amidst a floating neon jungle on a distant alien planet."*
![sample](./sample_1.png)
> *"A nostalgic 1950s oil painting of an EC4M woman reading a leather-bound book in a cozy, sun-drenched library filled with dust motes."*
![sample](./sample_2.png)
> *"A high-fashion editorial photo of an EC4M woman draped in iridescent silk, posing against the stark, brutalist concrete architecture of a futuristic city."*
## Use it with diffusers
```py
import torch
from diffusers import Krea2Pipeline
pipe = Krea2Pipeline.from_pretrained("krea/Krea-2-Turbo", torch_dtype=torch.bfloat16).to("cuda")
pipe.load_lora_weights("aimalias/ec4m")
image = pipe("A cinematic shot of an EC4M woman wearing bioluminescent armor, standing amidst a floating neon jungle on a distant alien planet.", num_inference_steps=8, guidance_scale=0.0).images[0]
image.save("output.png")
```