Module stable_diffusion 

Stable Diffusion

Stable Diffusion is a latent text-to-image diffusion model capable of generating photo-realistic images given any text input.

• 💻 Original Repository
• 🤗 Hugging Face
• The default scheduler for the v1.5, v2.1 and XL 1.0 version is the Denoising Diffusion Implicit Model scheduler (DDIM). The original paper and some code can be found in the associated repo. The default scheduler for the XL Turbo version is the Euler Ancestral scheduler.

Example

“A rusty robot holding a fire torch in its hand.” Generated by Stable Diffusion XL using Rust and candle.

# example running with cuda
# see the candle-examples/examples/stable-diffusion for all options
cargo run --example stable-diffusion --release --features=cuda,cudnn \
    -- --prompt "a cosmonaut on a horse (hd, realistic, high-def)"

# with sd-turbo
cargo run --example stable-diffusion --release --features=cuda,cudnn \
    -- --prompt "a cosmonaut on a horse (hd, realistic, high-def)" \
    --sd-version turbo

# with flash attention.
# feature flag: `--features flash-attn`
# cli flag: `--use-flash-attn`.
# flash-attention-v2 is only compatible with Ampere, Ada, \
# or Hopper GPUs (e.g., A100/H100, RTX 3090/4090).
cargo run --example stable-diffusion --release --features=cuda,cudnn \
    -- --prompt "a cosmonaut on a horse (hd, realistic, high-def)" \
    --use-flash-attn

Modules

attention

Attention Based Building Blocks

clip

Contrastive Language-Image Pre-Training

ddim

Denoising Diffusion Implicit Models

ddpm

embeddings

euler_ancestral_discrete

Ancestral sampling with Euler method steps.

resnet

ResNet Building Blocks

schedulers

Diffusion pipelines and models

unet_2d

2D UNet Denoising Models

unet_2d_blocks

2D UNet Building Blocks

uni_pc

UniPC Scheduler

utils

vae

Variational Auto-Encoder (VAE) Models.

Structs

StableDiffusionConfig

Functions

build_clip_transformer