mirror of https://github.com/leejet/stable-diffusion.cpp synced 2026-03-02 21:19:48 +01:00

Go to file

leejet f532972d60 fix: avoid precision issues on vulkan backend (#980 )		2025-11-16 20:57:08 +08:00
.github	chrore: enable Windows ROCm(HIP) build release (#956 )	2025-11-12 00:28:55 +08:00
assets	feat: add chroma radiance support (#910 )	2025-10-25 23:56:14 +08:00
docs	feat: support applying LoRA at runtime (#969 )	2025-11-13 21:48:44 +08:00
examples	feat: support independent sampler rng (#978 )	2025-11-16 17:11:02 +08:00
ggml@2d3876d554	sync: update ggml	2025-10-24 00:32:45 +08:00
thirdparty	feat: add wan2.1/2.2 support (#778 )	2025-09-06 18:08:03 +08:00
.clang-format	style: format code	2024-01-29 23:05:18 +08:00
.clang-tidy	chore: add .clang-tidy configuration and apply modernize checks (#902 )	2025-10-18 23:23:40 +08:00
.dockerignore	chore: add sd Dockerfile	2023-08-22 22:14:20 +08:00
.gitignore	feat: add image preview support (#522 )	2025-11-10 00:12:02 +08:00
.gitmodules	feat: add wan2.1/2.2 support (#778 )	2025-09-06 18:08:03 +08:00
clip.hpp	feat: support applying LoRA at runtime (#969 )	2025-11-13 21:48:44 +08:00
CMakeLists.txt	chore: allow building ggml as a separate shared lib (#468 )	2025-10-15 22:10:26 +08:00
common.hpp	fix: avoid precision issues on vulkan backend (#980 )	2025-11-16 20:57:08 +08:00
conditioner.hpp	feat: add break pseudo token support (#422 )	2025-11-16 14:45:20 +08:00
control.hpp	feat: support applying LoRA at runtime (#969 )	2025-11-13 21:48:44 +08:00
denoiser.hpp	refactor: unify the naming style of ggml extension functions (#921 )	2025-10-28 23:26:48 +08:00
diffusion_model.hpp	feat: support applying LoRA at runtime (#969 )	2025-11-13 21:48:44 +08:00
Dockerfile	chore: fix dockerfile libgomp1 dependency + improvements (#852 )	2025-10-11 00:17:45 +08:00
Dockerfile.musa	feat: upgrade musa sdk to rc4.2.0 (#732 )	2025-07-28 21:51:11 +08:00
Dockerfile.sycl	feat: Add SYCL Dockerfile (#651 )	2025-09-14 13:02:59 +08:00
esrgan.hpp	feat: support applying LoRA at runtime (#969 )	2025-11-13 21:48:44 +08:00
face_detect.py	feat: add PhotoMaker Version 2 support (#358 )	2024-11-23 11:50:14 +08:00
flux.hpp	feat: support applying LoRA at runtime (#969 )	2025-11-13 21:48:44 +08:00
format-code.sh	chore: add .clang-tidy configuration and apply modernize checks (#902 )	2025-10-18 23:23:40 +08:00
ggml_extend.hpp	feat: support applying LoRA at runtime (#969 )	2025-11-13 21:48:44 +08:00
gguf_reader.hpp	feat: add wan2.1/2.2 support (#778 )	2025-09-06 18:08:03 +08:00
gits_noise.inl	fix: clang complains about needless braces (#618 )	2025-03-09 12:26:41 +08:00
latent-preview.h	feat: add image preview support (#522 )	2025-11-10 00:12:02 +08:00
LICENSE	Initial commit	2023-08-13 16:00:22 +08:00
lora.hpp	feat: support applying LoRA at runtime (#969 )	2025-11-13 21:48:44 +08:00
ltxv.hpp	refactor: introduce GGMLRunnerContext (#928 )	2025-11-02 02:11:04 +08:00
mmdit.hpp	feat: support applying LoRA at runtime (#969 )	2025-11-13 21:48:44 +08:00
model.cpp	feat: support for --tensor-type-rules on generation modes (#932 )	2025-11-16 17:07:32 +08:00
model.h	feat: support for --tensor-type-rules on generation modes (#932 )	2025-11-16 17:07:32 +08:00
name_conversion.cpp	feat: support applying LoRA at runtime (#969 )	2025-11-13 21:48:44 +08:00
name_conversion.h	feat: support applying LoRA at runtime (#969 )	2025-11-13 21:48:44 +08:00
ordered_map.hpp	refactor: optimize the logic for name conversion and the processing of the LoRA model (#955 )	2025-11-10 00:12:20 +08:00
pmid.hpp	refactor: optimize the logic for name conversion and the processing of the LoRA model (#955 )	2025-11-10 00:12:20 +08:00
preprocessing.hpp	refactor: unify the naming style of ggml extension functions (#921 )	2025-10-28 23:26:48 +08:00
qwen_image.hpp	fix: avoid precision issues on vulkan backend (#980 )	2025-11-16 20:57:08 +08:00
qwenvl.hpp	feat: support applying LoRA at runtime (#969 )	2025-11-13 21:48:44 +08:00
README.md	feat: add cpu rng (#977 )	2025-11-16 14:48:15 +08:00
rng_mt19937.hpp	feat: add cpu rng (#977 )	2025-11-16 14:48:15 +08:00
rng_philox.hpp	chore: add .clang-tidy configuration and apply modernize checks (#902 )	2025-10-18 23:23:40 +08:00
rng.hpp	chore: add .clang-tidy configuration and apply modernize checks (#902 )	2025-10-18 23:23:40 +08:00
rope.hpp	refactor: introduce GGMLRunnerContext (#928 )	2025-11-02 02:11:04 +08:00
stable-diffusion.cpp	feat: support independent sampler rng (#978 )	2025-11-16 17:11:02 +08:00
stable-diffusion.h	feat: support independent sampler rng (#978 )	2025-11-16 17:11:02 +08:00
t5.hpp	refactor: optimize the logic for name conversion and the processing of the LoRA model (#955 )	2025-11-10 00:12:20 +08:00
tae.hpp	refactor: optimize the logic for name conversion and the processing of the LoRA model (#955 )	2025-11-10 00:12:20 +08:00
tokenize_util.cpp	feat: add Qwen Image support (#851 )	2025-10-12 23:23:19 +08:00
tokenize_util.h	feat: add Qwen Image support (#851 )	2025-10-12 23:23:19 +08:00
unet.hpp	feat: support applying LoRA at runtime (#969 )	2025-11-13 21:48:44 +08:00
upscaler.cpp	refactor: optimize the logic for name conversion and the processing of the LoRA model (#955 )	2025-11-10 00:12:20 +08:00
util.cpp	feat: add break pseudo token support (#422 )	2025-11-16 14:45:20 +08:00
util.h	feat: add image preview support (#522 )	2025-11-10 00:12:02 +08:00
vae.hpp	fix: compatibility for models with modified tensor shapes (#951 )	2025-11-07 23:04:41 +08:00
vocab_qwen.hpp	feat: add Qwen Image support (#851 )	2025-10-12 23:23:19 +08:00
vocab_umt5.hpp	feat: add wan2.1/2.2 support (#778 )	2025-09-06 18:08:03 +08:00
vocab.hpp	feat: add sd3 support (#298 )	2024-07-28 15:44:08 +08:00
wan.hpp	feat: support applying LoRA at runtime (#969 )	2025-11-13 21:48:44 +08:00

README.md

stable-diffusion.cpp

Diffusion model(SD,Flux,Wan,...) inference in pure C/C++

Note that this project is under active development.
API and command-line option may change frequently.

🔥Important News

2025/10/13 🚀 stable-diffusion.cpp now supports Qwen-Image-Edit / Qwen-Image-Edit 2509
👉 Details: PR #877
2025/10/12 🚀 stable-diffusion.cpp now supports Qwen-Image
👉 Details: PR #851
2025/09/14 🚀 stable-diffusion.cpp now supports Wan2.1 Vace
👉 Details: PR #819
2025/09/06 🚀 stable-diffusion.cpp now supports Wan2.1 / Wan2.2
👉 Details: PR #778

Features

Plain C/C++ implementation based on ggml, working in the same way as llama.cpp
Super lightweight and without external dependencies
Supported models
- Image Models
  - SD1.x, SD2.x, SD-Turbo
  - SDXL, SDXL-Turbo
  - Some SD1.x and SDXL distilled models
  - SD3/SD3.5
  - Flux-dev/Flux-schnell
  - Chroma
  - Chroma1-Radiance
  - Qwen Image
- Image Edit Models
  - FLUX.1-Kontext-dev
  - Qwen Image Edit/Qwen Image Edit 2509
- Video Models
  - Wan2.1/Wan2.2
- PhotoMaker support.
- Control Net support with SD 1.5
- LoRA support, same as stable-diffusion-webui
- Latent Consistency Models support (LCM/LCM-LoRA)
- Faster and memory efficient latent decoding with TAESD
- Upscale images generated with ESRGAN
Supported backends
- CPU (AVX, AVX2 and AVX512 support for x86 architectures)
- CUDA
- Vulkan
- Metal
- OpenCL
- SYCL
Supported weight formats
- Pytorch checkpoint (.ckpt or .pth)
- Safetensors (./safetensors)
- GGUF (.gguf)
Supported platforms
- Linux
- Mac OS
- Windows
- Android (via Termux, Local Diffusion)
Flash Attention for memory usage optimization
Negative prompt
stable-diffusion-webui style tokenizer (not all the features, only token weighting for now)
VAE tiling processing for reduce memory usage
Sampling method
- Euler A
- Euler
- Heun
- DPM2
- DPM++ 2M
- DPM++ 2M v2
- DPM++ 2S a
- LCM
Cross-platform reproducibility
- --rng cuda, default, consistent with the stable-diffusion-webui GPU RNG
- --rng cpu, consistent with the comfyui RNG
Embedds generation parameters into png output as webui-compatible text string

Quick Start

Get the sd executable

Download pre-built binaries from the releases page
Or build from source by following the build guide

Download model weights

download weights(.ckpt or .safetensors or .gguf). For example
- Stable Diffusion v1.5 from https://huggingface.co/runwayml/stable-diffusion-v1-5
```
curl -L -O https://huggingface.co/runwayml/stable-diffusion-v1-5/resolve/main/v1-5-pruned-emaonly.safetensors
```

Generate an image with just one command

./bin/sd -m ../models/v1-5-pruned-emaonly.safetensors -p "a lovely cat"

For detailed command-line arguments, check out cli doc.

Performance

If you want to improve performance or reduce VRAM/RAM usage, please refer to performance guide.

More Guides

Bindings

These projects wrap stable-diffusion.cpp for easier use in other languages/frameworks.

Golang (non-cgo): seasonjs/stable-diffusion
Golang (cgo): Binozo/GoStableDiffusion
C#: DarthAffe/StableDiffusion.NET
Python: william-murray1204/stable-diffusion-cpp-python
Rust: newfla/diffusion-rs
Flutter/Dart: rmatif/Local-Diffusion

UIs

These projects use stable-diffusion.cpp as a backend for their image generation.

Contributors

Thank you to all the people who have already contributed to stable-diffusion.cpp!