Download
A CLI tool/python module for generating images from text using guided diffusion and CLIP from OpenAI. Text to image generation (multiple prompts with weights). Non-square Generations (experimental) Generate portrait or landscape images by specifying a number to offset the width and/or height. Uses fewer timesteps over the same diffusion schedule. Sacrifices accuracy/alignment for quicker runtime. options: - 25, 50, 150, 250, 500, 1000, ddim25,ddim50,ddim150, ddim250,ddim500,ddim1000 (default: 1000) Prepending a number with ddim will use the ddim scheduler. e.g. ddim25 will use the 25 timstep ddim scheduler. This method may be better at shorter timestep_respacing values. Multiple prompts can be specified with the | character. You may optionally specify a weight for each prompt.
Features
- Text to image generation
- Text to image generation (multiple prompts with weights)
- Iterations/Steps (Timestep Respacing)
- Blend an image with the diffusion for a number of steps
- Non-square Generations (experimental) Generate portrait or landscape images by specifying a number to offset the width and/or height
- Some tests require a GPU
Project Samples
