Ecosyste.ms: Awesome

An open API service indexing awesome lists of open source software.

Awesome Lists | Featured Topics | Projects

https://github.com/oracle/sd4j

Stable diffusion pipeline in Java using ONNX Runtime
https://github.com/oracle/sd4j

java onnxruntime stable-diffusion

Last synced: 11 days ago
JSON representation

Stable diffusion pipeline in Java using ONNX Runtime

Awesome Lists containing this project

README 

        
# SD4J (Stable Diffusion in Java)



This repo contains an implementation of Stable Diffusion inference running on top of ONNX Runtime,

written in Java. It's a modified port of the [C# implementation](https://github.com/cassiebreviu/StableDiffusion/),

with a GUI for repeated generations and support for negative text inputs. It is intended to be a

demonstration of how to use ONNX Runtime from Java, and best practices for ONNX Runtime to get good performance.

We will keep it current with the latest releases of ONNX Runtime, with appropriate updates as new performance

related ONNX Runtime features become available through the ONNX Runtime Java API. All the code is subject to change as

this is a code sample, any APIs in it should not be considered stable.



This repo targets ONNX Runtime 1.14. The version number is in two parts `-`, and the

initial release of sd4j is v1.0-1.14.0. We'll bump the sd4j version number if it gains new features and the ONNX Runtime

version number as we depend on newer versions of ONNX Runtime.



The project supports txt2img generation, it doesn't currently implement img2img, upscaling or inpainting.



By default it uses a fp32 model, and running on a 6 core 2019 16" Intel Macbook Pro each diffusion step takes around 5s. 

Running on better hardware, or with a CUDA GPU will greatly reduce the time taken to generate an image, as will using an

SD-Turbo model. There is experimental support for the CoreML (for macOS) and DirectML (for Windows) backends, but proper 

utilisation of these may require model changes like quantization which is not yet implemented.



## Example images



These are a few example images generated by this code along with their generation parameters:



![Generated image from the prompt "Wildlife photograph of an astronaut riding a horse in the desert"](images/astronaut-horse.png "Wildlife photograph of an astronaut riding a horse in the desert")



Text: "Wildlife photograph of an astronaut riding a horse in the desert", Negative Text: "", Seed: 42, Guidance Scale: 10, Inference Steps: 40, Scheduler: Euler Ancestral, Image Size: 512x512.



![Generated image from the prompt "Press photo of an America's Cup catamaran sailing through the sands of Mars, high resolution, high quality"](images/boat-mars.png "Press photo of an America's Cup catamaran sailing through the sands of Mars")



Text: "Press photo of an America's Cup catamaran sailing through the sands of Mars, high resolution, high quality", Negative Text: "water, sea, ocean, lake", Seed: 42, Guidance Scale: 10, Inference Steps: 40, Scheduler: Euler Ancestral, Image Size: 512x512.



![Generated image from the prompt "Professional photograph of the Apollo 11 lunar lander in a field, high quality, 4k"](images/lunar-lander.png "Professional photograph of the Apollo 11 lunar lander in a field, high quality, 4k")



Text: "Professional photograph of the Apollo 11 lunar lander in a field, high quality, 4k", Negative Text: "", Seed: 42, Guidance Scale: 10, Inference Steps: 50, Scheduler: Euler Ancestral, Image Size: 512x512.



![Generated image from the prompt "Professional photograph of George Washington in his garden grilling steaks, detailed face, high quality, 4k"](images/washington-steak.png "Professional photograph of George Washington in his garden grilling steaks, detailed face, high quality, 4k")



Text: "Professional photograph of George Washington in his garden grilling steaks, detailed face, high quality, 4k", Negative Text: "painting, drawing, art", Seed: 42, Guidance Scale: 10, Inference Steps: 60, Scheduler: Euler Ancestral, Image Size: 512x512.



## Model support



The SD4J project supports SD v1.5, SD v2 and SDXL style models. For models which do not support classifier-free guidance

or negative prompts, such as SD-Turbo or SDXL-Turbo, the guidance scale should be set to a value less than 1.0 which

disables that guidance. Models like SD-Turbo can generate acceptable images in as few as two diffusion steps. The 

difference between SDv1 and SDv2 models is autodetected, but SDXL must be supplied as the model type for SDXL models

otherwise it will throw an exception on generation. In some cases the autodetection of v1 and v2 may fail in which case

supplying the `--model-type {SD1.5, SD2, SDXL}` argument with the appropriate parameter will fix the model type.



## Installation



This project requires [Apache Maven](https://maven.apache.org), [Java 17 or newer](https://www.oracle.com/java/technologies/downloads/),

a compiled ONNX Runtime extensions binary, and a Stable Diffusion model checkpoint.

The other dependencies (ONNX Runtime and Apache Commons Math) are downloaded by Maven automatically.



### Prepare model checkpoint



There are many compatible models on [Hugging Face's website](https://huggingface.co). We have tested the

Stable Diffusion v1.5 checkpoint, which has pre-built ONNX models. This can be downloaded via 

the following `git` commands (skip the first one if you have already configured `git-lfs`):

```bash

git lfs install

git clone https://huggingface.co/runwayml/stable-diffusion-v1-5 -b onnx

```

The Stable Diffusion v1.5 checkpoint is available under the [OpenRAIL-M license](https://github.com/CompVis/stable-diffusion/blob/main/LICENSE).

For other SD models there is a one or two stage process to generate the ONNX format models. If the model is already in 

Hugging Face Diffusers format then you can run the `convert_stable_diffusion_checkpoint_to_onnx.py` file from the 

[diffusers](https://github.com/huggingface/diffusers) project as follows:

```bash

python scripts/convert_stable_diffusion_checkpoint_to_onnx.py --model_path  --output_path 

```

If the model is an original stable diffusion checkpoint then you first need to run:

```bash

python scripts/convert_original_stable_diffusion_to_diffusers.py --checkpoint_path  --scheduler_type lms --dump_path 

```

If the model is an SDXL model then it needs to be exported from the Hugging Face Hub 

using [optimum](https://github.com/huggingface/optimum):

```bash

optimum-cli export onnx --model  

```

The scripts require a suitable Python 3 virtual environment with `diffusers`, `onnxruntime`, `optimum` and `onnx` 

installed.



### Setup ORT extensions

You will also need to check out and compile onnxruntime-extensions for your platform. The repo is [https://github.com/microsoft/onnxruntime-extensions](https://github.com/microsoft/onnxruntime-extensions),

and it can be compiled with `./build_lib.sh --config Release --update --build --parallel` which generates the required library (`libortextensions.[dylib,so]` or `ortextensions.dll`) in the

`build//Release/lib/` folder. That library should be copied into the root of this directory.



## Running the GUI



The GUI can be executed with `mvn package exec:exec -DmodelPath=`. It constructs a

window where you can specify the parameters of the image you'd like to generate, and each image creates its own window

where it can save the image as a png file. Saved png files contain a metadata field with the generation parameters.



### Use in other programs



The `com.oracle.labs.mlrg.sd4j.SD4J` class provides a full image generation pipeline which can be used without the GUI

directly from other code.



### Using a CUDA GPU



To use the GPU you need to modify the pom file to depend on `onnxruntime_gpu` and swap `CPU` for

`CUDA` in the `exec-maven-plugin` block.



## Implementation details



This code provides a thin `Tensor` wrapper object which is a tuple of a direct `ByteBuffer` instance and a long shape

array, which is used to provide easy access in and out of ORT's `OnnxTensor` objects. There's a `Scheduler` interface 

which the two available schedulers (LMS and Euler Ancestral) implement. The `SD4J` pipeline object is a suitable entry

point for use without the GUI, and there is an example of such usage in the `CLIApp` class.



## Contributing



This project welcomes contributions from the community. Before submitting a pull request, please [review our contribution guide](./CONTRIBUTING.md).



## Security



Please consult the [security guide](./SECURITY.md) for our responsible security vulnerability disclosure process



## License



The code is available under the [Universal Permissive License (UPL)](https://oss.oracle.com/licenses/upl/). It requires

a Stable Diffusion model architecture checkpoint to work, and any Stable Diffusion models should be used under their

licenses. There are 1000+ compatible models available

on [Hugging Face](https://huggingface.co/models?other=stable-diffusion) each of which are licensed separately, though many use a variant of

the [OpenRAIL-M license](https://github.com/CompVis/stable-diffusion/blob/main/LICENSE).



The [tokenizer onnx model](text_tokenizer/custom_op_cliptok.onnx) is taken from the 

[C# implementation](https://github.com/cassiebreviu/StableDiffusion/), and is available under the MIT license. More 

details on the tokenizer are available in its [README file](text_tokenizer/README.md).