[Announcement] Some txt2img options will be renamed soon!
fboulnois opened this issue · 0 comments
fboulnois commented
The original txt2img.py
script is rapidly becoming outdated and some of the options from txt2img.py
are hard to remember. Since then, stable-diffusion-docker
has continued to add newer pipelines from diffusers
like onnx
, upscale4x
and inpaint
, add newer features like negative prompts and attention slicing, support newer models like stable diffusion 1.5, 2.0, and 2.1, and simplify and standardize options where possible.
In the future, some of the original txt2img
options will be renamed for easy-of-use and compatibility with other tools and pipelines:
txt2img | stable-diffusion-docker |
---|---|
--H |
--height |
--W |
--width |
--n_iter |
--iters |
--n_samples |
--samples |
--ddim_steps |
--steps |