Text-to-Image
Diffusers
Safetensors
English
StableDiffusionPipeline
art
stable-diffusion
pokemon
fine-tuned
non-commercial
research
import torch
from diffusers import DiffusionPipeline
# switch to "mps" for apple devices
pipe = DiffusionPipeline.from_pretrained("PPPPPilot/StableDiffusionPokemon", dtype=torch.bfloat16, device_map="cuda")
prompt = "Astronaut in a jungle, cold color palette, muted colors, detailed, 8k"
image = pipe(prompt).images[0]🔬 Research Use | Non-Commercial
📄 License
This model is fine-tuned from sd-legacy/stable-diffusion-v1-5 under CreativeML Open RAIL-M License.
🚀 Usage
from diffusers import StableDiffusionPipeline
pipe = StableDiffusionPipeline.from_pretrained("PPPPPilot/StableDiffusionPokemon")
image = pipe("a cute pokemon character").images[0]
image.save("my_pokemon.png")
✨ Example Display
Input Prompt: "a cute pokemon character"
Input Prompt: "a green color frog pokemon character"
Input Prompt: "a singer pokemon character looks like the American Singer star Taylor Swift"
Input Prompt: "a pokemon character looks like Chinese action movie star Jackie Chan"
Input Prompt: "a US President Donald Trump pokemon character, human-like"
✍🏼 Technical Specification
Dataset: reach-vb/pokemon-blip-captions
Training Steps: 10,000
Evaluation Score: FID 60.20, CLIP 0.235
License: CreativeML Open RAIL-M (Non-commercial)
- Downloads last month
- 5
Model tree for PPPPPilot/StableDiffusionPokemon
Base model
stable-diffusion-v1-5/stable-diffusion-v1-5