SDXL LoRA DreamBooth - computational-mama/underwater-humanoid

Prompt

A <s0><s1> character, An underwater humanoid that floats surrounded by coral, fish and algae, with translucent skin, no genus, no nose, a soft body, large fish-like antennae on the face, and tentacular, webbed extremities. In the background, a submarine cave lets in light from the surface. Photorealistic

Prompt

A <s0><s1> character, An underwater humanoid with salt crystals on its face floating in a seated position surrounded by fish, coral and algae, with translucent skin, no genus, no nose, veins and organs visible through the skin, spaced-aperture eyes without a globe, and webbed extremities. In the background, a submarine cave lets in light from the surface. Photorealistic

Prompt

A <s0><s1> character, An underwater humanoid-algae that swims surrounded by coral and fish and algae, harvests algae, with translucent skin, no genus, no nose, organs visible through the skin, it has a head of large fish antennae on its face, and tentacular, webbed extremities. Fish are attached to its body. In the background, a submarine cave lets in light from the surface. Photorealistic

Prompt

A <s0><s1> character, A winged underwater humanoid sits surrounded by its garden of coral and algae, with translucent, genderless skin, a snout-like nose, muscles and organs visible through the skin, large globe-shaped eyes and webbed extremities. In the background, a submarine cave lets in light from the surface. Photorealistic

Prompt

A <s0><s1> character, An underwater humanoid crowned with translucent antennae appears from behind corals and algae, with translucent skin, no genus, no nose, muscles and organs visible through the skin, flexible, tubular limbs, no eyes, and slender, humanoid extremities. In the background, a submarine cave lets in light from the surface. Photorealistic

Model description

These are computational-mama/underwater-humanoid LoRA adaption weights for stabilityai/stable-diffusion-xl-base-1.0.

Download model

Use it with UIs such as AUTOMATIC1111, Comfy UI, SD.Next, Invoke

• LoRA: download underwater-humanoid.safetensors here 💾.
  • Place it on your models/Lora folder.
  • On AUTOMATIC1111, load the LoRA by adding <lora:underwater-humanoid:1> to your prompt. On ComfyUI just load it as a regular LoRA.
• Embeddings: download underwater-humanoid_emb.safetensors here 💾.
  • Place it on it on your embeddings folder
  • Use it by adding underwater-humanoid_emb to your prompt. For example, A underwater-humanoid_emb character (you need both the LoRA and the embeddings as they were trained together for this LoRA)

Use it with the 🧨 diffusers library

from diffusers import AutoPipelineForText2Image
import torch
from huggingface_hub import hf_hub_download
from safetensors.torch import load_file
        
pipeline = AutoPipelineForText2Image.from_pretrained('stabilityai/stable-diffusion-xl-base-1.0', torch_dtype=torch.float16).to('cuda')
pipeline.load_lora_weights('computational-mama/underwater-humanoid', weight_name='pytorch_lora_weights.safetensors')
embedding_path = hf_hub_download(repo_id='computational-mama/underwater-humanoid', filename='underwater-humanoid_emb.safetensors' repo_type="model")
state_dict = load_file(embedding_path)
pipeline.load_textual_inversion(state_dict["clip_l"], token=["<s0>", "<s1>"], text_encoder=pipeline.text_encoder, tokenizer=pipeline.tokenizer)
pipeline.load_textual_inversion(state_dict["clip_g"], token=["<s0>", "<s1>"], text_encoder=pipeline.text_encoder_2, tokenizer=pipeline.tokenizer_2)
        
image = pipeline('A <s0><s1> character').images[0]

For more details, including weighting, merging and fusing LoRAs, check the documentation on loading LoRAs in diffusers

Trigger words

To trigger image generation of trained concept(or concepts) replace each concept identifier in you prompt with the new inserted tokens:

to trigger concept TOK → use <s0><s1> in your prompt

Details

All Files & versions.

The weights were trained using 🧨 diffusers Advanced Dreambooth Training Script.

LoRA for the text encoder was enabled. False.

Pivotal tuning was enabled: True.

Special VAE used for training: madebyollin/sdxl-vae-fp16-fix.