Instructions to use Elvenson/diffuser_inference with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Diffusers
How to use Elvenson/diffuser_inference with Diffusers:
pip install -U diffusers transformers accelerate
import torch from diffusers import DiffusionPipeline # switch to "mps" for apple devices pipe = DiffusionPipeline.from_pretrained("Elvenson/diffuser_inference", dtype=torch.bfloat16, device_map="cuda") prompt = "Astronaut in a jungle, cold color palette, muted colors, detailed, 8k" image = pipe(prompt).images[0] - Notebooks
- Google Colab
- Kaggle
| import argparse | |
| import base64 | |
| from io import BytesIO | |
| from PIL import Image | |
| from handler import EndpointHandler, decode_base64_image | |
| def local_predict(prompts, encode_image): | |
| # Init handler | |
| my_handler = EndpointHandler() | |
| if encode_image: | |
| response = my_handler({"inputs": prompts, "image": encode_image}) | |
| else: | |
| response = my_handler({"inputs": prompts}) | |
| image = decode_base64_image(response["image"]) | |
| image.save("local_output.png") | |
| opt = argparse.ArgumentParser("Diffuser local test") | |
| opt.add_argument("-prompts", "--prompts", default="", type=str, help="Diffuser prompts") | |
| opt.add_argument("-image", "--image", default="", type=str, help="Init image") | |
| if __name__ == '__main__': | |
| args = opt.parse_args() | |
| encoded_string = "" | |
| if args.image: | |
| with open(args.image, "rb") as image_file: | |
| encoded_string = base64.b64encode(image_file.read()).decode() | |
| local_predict(args.prompts, encoded_string) | |