Expand description
Local image generation backend for Blazen using diffusion-rs.
This crate wraps the diffusion-rs pure-Rust Stable Diffusion inference
engine to provide fully local, offline image generation with no API keys
required.
When used through blazen-llm with the diffusion feature flag, this
crate’s DiffusionProvider will implement blazen_llm::ImageGeneration.
§Feature flags
| Feature | Description |
|---|---|
engine | Links the actual diffusion-rs runtime (CPU) |
cuda | NVIDIA CUDA GPU acceleration |
metal | Apple Silicon GPU acceleration (Metal) |
Without the engine feature the crate compiles (options struct + stub
provider) but cannot actually run image generation. This keeps workspace
builds fast when the heavy native dependencies are not needed.
Structs§
- Diffusion
Options - Options for constructing a
DiffusionProvider. - Diffusion
Provider - A local image generation provider backed by
diffusion-rs.
Enums§
- Diffusion
Error - Error type for diffusion-rs operations.
- Diffusion
Scheduler - Noise schedulers available for the diffusion process.