--- base_model: krea/Krea-2-Raw tags: - text-to-image - diffusers - lora - krea2 - template:sd-lora license: apache-2.0 instance_prompt: "B3MM4 woman" widget: - text: "A cinematic shot of a B3MM4 woman dressed in ornate gold armor, standing atop a floating crystal peak amidst a swirling nebula of violet and teal gas." output: url: sample_0.png - text: "A candid, grainy 35mm film photograph of a B3MM4 woman reading a vintage book in a sun-drenched Parisian cafe with steam rising from a porcelain cup." output: url: sample_1.png - text: "A hyper-realistic portrait of a B3MM4 woman as a cyberpunk hacker, illuminated by flickering neon holographic screens in a rain-slicked midnight alleyway." output: url: sample_2.png --- # Krea 2 LoRA — aimalias/b3mm4 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 `B3MM4 woman` to invoke the concept. ## Samples ![sample](./sample_0.png) > *"A cinematic shot of a B3MM4 woman dressed in ornate gold armor, standing atop a floating crystal peak amidst a swirling nebula of violet and teal gas."* ![sample](./sample_1.png) > *"A candid, grainy 35mm film photograph of a B3MM4 woman reading a vintage book in a sun-drenched Parisian cafe with steam rising from a porcelain cup."* ![sample](./sample_2.png) > *"A hyper-realistic portrait of a B3MM4 woman as a cyberpunk hacker, illuminated by flickering neon holographic screens in a rain-slicked midnight alleyway."* ## 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/b3mm4") image = pipe("A cinematic shot of a B3MM4 woman dressed in ornate gold armor, standing atop a floating crystal peak amidst a swirling nebula of violet and teal gas.", num_inference_steps=8, guidance_scale=0.0).images[0] image.save("output.png") ```