Skip to content

Features

Jump to bottom
cmdr2 edited this page Mar 11, 2023 · 14 revisions

t2i

Supports: Windows, Linux and Mac.

Meta

  • A clean and simple API and implementation, to allow adding new features easily.
  • A single, distributable package that can be installed with pip. Conda is not mandatory. pip install sdkit will automatically install all the dependencies (with the exception of torch and torchvision, since they don't host the CUDA versions on PyPI).

Image Generation

  • Stable Diffusion 2.1 support.
  • txt2img.
  • img2img.
  • Inpainting.
  • 19 Samplers: ddim, plms, heun, euler, euler_a, dpm2, dpm2_a, lms, dpm_solver_stability, dpmpp_2s_a, dpmpp_2m, dpmpp_sde, dpm_fast, dpm_adaptive, unipc_snr, unipc_tu, unipc_tq, unipc_snr_2, unipc_tu_2.
  • Negative Prompt: Specify aspects of the image to remove. Supports weights.
  • Attention/Emphasis: () in the prompt increases the model's attention to enclosed words, and [] decreases it.
  • Weighted Prompts: Use weights for specific words in your prompt to change their importance, e.g. red:2.4 dragon:1.2.
  • Live Preview: Support for image callbacks, to get the images as they are being generated.
  • Save to disk: Built-in functions to save images and metadata to disk.

Image Filters

  • GFPGAN for face restoration.
  • RealESRGAN (and anime) for upscaling. Supports 2x and 4x upscaling.
  • NSFW Checker for detecting (and blurring) potentially unsafe images.

Models

  • Models DB lets you easily download known models with a single line of code.
  • Models DB allows automatic detection of the config file to use (e.g. for SD 2.1 and 2.0 "v" type models).
  • Fast loading/unloading of VAE models, without needing to reload the entire Stable Diffusion model.
  • Custom models for Stable Diffusion in .ckpt or .safetensors format.
  • Custom models for VAE in .ckpt or .safetensors format.
  • Custom models for Hypernetworks in .pt or .safetensors format.
  • API to merge two models (in .ckpt or .safetensors format).

Performance & Security

  • VRAM Optimizations to allow generating 512x512 images on GPUs with less than 4gb of VRAM.
  • "Balanced" optimizations to achieve high rendering speed with fairly low VRAM usage.
  • Support for .safetensors file format, for secure model loading.
  • Automatic scanning of Stable Diffusion models before loading them (picklescan).
  • Support for running on multiple GPUs in parallel.

Upcoming features/roadmap

📢 We're looking for code contributions for these features (or anything else you'd like to work on):

  • CodeFormer upscaling (please maintain the required copyright notices)
  • Using custom Textual Inversion embeddings
  • Seamless tiling
  • Outpainting
  • AMD support
  • Allow other samplers for img2img (instead of only DDIM)
Clone this wiki locally