xDiT diffusion inference#
2026-02-11
27 min read time
The rocm/pytorch-xdit Docker image offers a prebuilt, optimized environment based on xDiT for benchmarking diffusion model video and image generation on gfx942 and gfx950 series (AMD Instinct™ MI300X, MI325X, MI350X, and MI355X) GPUs. The image runs ROCm 7.11.0 (preview) based on TheRock and includes the following components:
Software components - xdit:v26.1
Software component |
Version |
|---|---|
1728a81 |
|
d23d18f |
|
ab0101c |
|
a2f7c35 |
|
659737c |
|
91be249 |
|
b919bd0 |
|
a272dfa |
|
b521400f |
|
de14bec0 |
|
6708f5 |
|
0a3d7a |
|
2c9b712 |
Follow this guide to pull the required image, spin up a container, download the model, and run a benchmark. For preview and development releases, see amdsiloai/pytorch-xdit.
What’s new#
HunyuanVideo 1.5 support
Z-Image Turbo support
Wan model sharding
Supported models#
The following models are supported for inference performance benchmarking. Some instructions, commands, and recommendations in this documentation might vary by model – select one to get started.
Note
To learn more about your specific model see the Hunyuan Video model card on Hugging Face or visit the GitHub page. Note that some models require access authorization before use via an external license agreement through a third party.
Note
To learn more about your specific model see the Hunyuan Video 1.5 model card on Hugging Face or visit the GitHub page. Note that some models require access authorization before use via an external license agreement through a third party.
Note
To learn more about your specific model see the Wan2.1 model card on Hugging Face or visit the GitHub page. Note that some models require access authorization before use via an external license agreement through a third party.
Note
To learn more about your specific model see the Wan2.2 model card on Hugging Face or visit the GitHub page. Note that some models require access authorization before use via an external license agreement through a third party.
Note
To learn more about your specific model see the FLUX.1 model card on Hugging Face or visit the GitHub page. Note that some models require access authorization before use via an external license agreement through a third party.
Note
To learn more about your specific model see the FLUX.1 Kontext model card on Hugging Face or visit the GitHub page. Note that some models require access authorization before use via an external license agreement through a third party.
Note
To learn more about your specific model see the FLUX.2 model card on Hugging Face or visit the GitHub page. Note that some models require access authorization before use via an external license agreement through a third party.
Note
To learn more about your specific model see the stable-diffusion-3.5-large model card on Hugging Face or visit the GitHub page. Note that some models require access authorization before use via an external license agreement through a third party.
Note
To learn more about your specific model see the Z-Image Turbo model card on Hugging Face or visit the GitHub page. Note that some models require access authorization before use via an external license agreement through a third party.
System validation#
Before running AI workloads, it’s important to validate that your AMD hardware is configured correctly and performing optimally.
If you have already validated your system settings, including aspects like NUMA auto-balancing, you can skip this step. Otherwise, complete the procedures in the System validation and optimization guide to properly configure your system settings before starting.
To test for optimal performance, consult the recommended System health benchmarks. This suite of tests will help you verify and fine-tune your system’s configuration.
Pull the Docker image#
For this tutorial, it’s recommended to use the latest rocm/pytorch-xdit:v26.1 Docker image.
Pull the image using the following command:
docker pull rocm/pytorch-xdit:v26.1
Validate and benchmark#
Once the image has been downloaded you can follow these steps to run benchmarks and generate outputs.
The following commands are written for Hunyuan Video. See Supported models to switch to another available model.
The following commands are written for Hunyuan Video 1.5. See Supported models to switch to another available model.
The following commands are written for Wan2.1. See Supported models to switch to another available model.
The following commands are written for Wan2.2. See Supported models to switch to another available model.
The following commands are written for FLUX.1. See Supported models to switch to another available model.
The following commands are written for FLUX.1 Kontext. See Supported models to switch to another available model.
The following commands are written for FLUX.2. See Supported models to switch to another available model.
The following commands are written for stable-diffusion-3.5-large. See Supported models to switch to another available model.
The following commands are written for Z-Image Turbo. See Supported models to switch to another available model.
Choose your setup method#
You can either use an existing Hugging Face cache or download the model fresh inside the container.
If you already have models downloaded on your host system, you can mount your existing cache.
Set your Hugging Face cache location.
export HF_HOME=/your/hf_cache/location
Download the model (if not already cached).
huggingface-cli download tencent/HunyuanVideo --revision refs/pr/18
Launch the container with mounted cache.
docker run \ -it --rm \ --cap-add=SYS_PTRACE \ --security-opt seccomp=unconfined \ --user root \ --device=/dev/kfd \ --device=/dev/dri \ --group-add video \ --ipc=host \ --network host \ --privileged \ --shm-size 128G \ --name pytorch-xdit \ -e HSA_NO_SCRATCH_RECLAIM=1 \ -e OMP_NUM_THREADS=16 \ -e CUDA_VISIBLE_DEVICES=0,1,2,3,4,5,6,7 \ -e HF_HOME=/app/huggingface_models \ -v $HF_HOME:/app/huggingface_models \ rocm/pytorch-xdit:v26.1
If you prefer to keep the container self-contained or don’t have an existing cache.
Launch the container
docker run \ -it --rm \ --cap-add=SYS_PTRACE \ --security-opt seccomp=unconfined \ --user root \ --device=/dev/kfd \ --device=/dev/dri \ --group-add video \ --ipc=host \ --network host \ --privileged \ --shm-size 128G \ --name pytorch-xdit \ -e HSA_NO_SCRATCH_RECLAIM=1 \ -e OMP_NUM_THREADS=16 \ -e CUDA_VISIBLE_DEVICES=0,1,2,3,4,5,6,7 \ rocm/pytorch-xdit:v26.1
Inside the container, set the Hugging Face cache location and download the model.
export HF_HOME=/app/huggingface_models huggingface-cli download tencent/HunyuanVideo --revision refs/pr/18
Warning
Models will be downloaded to the container’s filesystem and will be lost when the container is removed unless you persist the data with a volume.
If you already have models downloaded on your host system, you can mount your existing cache.
Set your Hugging Face cache location.
export HF_HOME=/your/hf_cache/location
Download the model (if not already cached).
huggingface-cli download hunyuanvideo-community/HunyuanVideo-1.5-Diffusers-720p_t2v
Launch the container with mounted cache.
docker run \ -it --rm \ --cap-add=SYS_PTRACE \ --security-opt seccomp=unconfined \ --user root \ --device=/dev/kfd \ --device=/dev/dri \ --group-add video \ --ipc=host \ --network host \ --privileged \ --shm-size 128G \ --name pytorch-xdit \ -e HSA_NO_SCRATCH_RECLAIM=1 \ -e OMP_NUM_THREADS=16 \ -e CUDA_VISIBLE_DEVICES=0,1,2,3,4,5,6,7 \ -e HF_HOME=/app/huggingface_models \ -v $HF_HOME:/app/huggingface_models \ rocm/pytorch-xdit:v26.1
If you prefer to keep the container self-contained or don’t have an existing cache.
Launch the container
docker run \ -it --rm \ --cap-add=SYS_PTRACE \ --security-opt seccomp=unconfined \ --user root \ --device=/dev/kfd \ --device=/dev/dri \ --group-add video \ --ipc=host \ --network host \ --privileged \ --shm-size 128G \ --name pytorch-xdit \ -e HSA_NO_SCRATCH_RECLAIM=1 \ -e OMP_NUM_THREADS=16 \ -e CUDA_VISIBLE_DEVICES=0,1,2,3,4,5,6,7 \ rocm/pytorch-xdit:v26.1
Inside the container, set the Hugging Face cache location and download the model.
export HF_HOME=/app/huggingface_models huggingface-cli download hunyuanvideo-community/HunyuanVideo-1.5-Diffusers-720p_t2v
Warning
Models will be downloaded to the container’s filesystem and will be lost when the container is removed unless you persist the data with a volume.
If you already have models downloaded on your host system, you can mount your existing cache.
Set your Hugging Face cache location.
export HF_HOME=/your/hf_cache/location
Download the model (if not already cached).
huggingface-cli download Wan-AI/Wan2.1-I2V-14B-720P-Diffusers
Launch the container with mounted cache.
docker run \ -it --rm \ --cap-add=SYS_PTRACE \ --security-opt seccomp=unconfined \ --user root \ --device=/dev/kfd \ --device=/dev/dri \ --group-add video \ --ipc=host \ --network host \ --privileged \ --shm-size 128G \ --name pytorch-xdit \ -e HSA_NO_SCRATCH_RECLAIM=1 \ -e OMP_NUM_THREADS=16 \ -e CUDA_VISIBLE_DEVICES=0,1,2,3,4,5,6,7 \ -e HF_HOME=/app/huggingface_models \ -v $HF_HOME:/app/huggingface_models \ rocm/pytorch-xdit:v26.1
If you prefer to keep the container self-contained or don’t have an existing cache.
Launch the container
docker run \ -it --rm \ --cap-add=SYS_PTRACE \ --security-opt seccomp=unconfined \ --user root \ --device=/dev/kfd \ --device=/dev/dri \ --group-add video \ --ipc=host \ --network host \ --privileged \ --shm-size 128G \ --name pytorch-xdit \ -e HSA_NO_SCRATCH_RECLAIM=1 \ -e OMP_NUM_THREADS=16 \ -e CUDA_VISIBLE_DEVICES=0,1,2,3,4,5,6,7 \ rocm/pytorch-xdit:v26.1
Inside the container, set the Hugging Face cache location and download the model.
export HF_HOME=/app/huggingface_models huggingface-cli download Wan-AI/Wan2.1-I2V-14B-720P-Diffusers
Warning
Models will be downloaded to the container’s filesystem and will be lost when the container is removed unless you persist the data with a volume.
If you already have models downloaded on your host system, you can mount your existing cache.
Set your Hugging Face cache location.
export HF_HOME=/your/hf_cache/location
Download the model (if not already cached).
huggingface-cli download Wan-AI/Wan2.2-I2V-A14B-Diffusers
Launch the container with mounted cache.
docker run \ -it --rm \ --cap-add=SYS_PTRACE \ --security-opt seccomp=unconfined \ --user root \ --device=/dev/kfd \ --device=/dev/dri \ --group-add video \ --ipc=host \ --network host \ --privileged \ --shm-size 128G \ --name pytorch-xdit \ -e HSA_NO_SCRATCH_RECLAIM=1 \ -e OMP_NUM_THREADS=16 \ -e CUDA_VISIBLE_DEVICES=0,1,2,3,4,5,6,7 \ -e HF_HOME=/app/huggingface_models \ -v $HF_HOME:/app/huggingface_models \ rocm/pytorch-xdit:v26.1
If you prefer to keep the container self-contained or don’t have an existing cache.
Launch the container
docker run \ -it --rm \ --cap-add=SYS_PTRACE \ --security-opt seccomp=unconfined \ --user root \ --device=/dev/kfd \ --device=/dev/dri \ --group-add video \ --ipc=host \ --network host \ --privileged \ --shm-size 128G \ --name pytorch-xdit \ -e HSA_NO_SCRATCH_RECLAIM=1 \ -e OMP_NUM_THREADS=16 \ -e CUDA_VISIBLE_DEVICES=0,1,2,3,4,5,6,7 \ rocm/pytorch-xdit:v26.1
Inside the container, set the Hugging Face cache location and download the model.
export HF_HOME=/app/huggingface_models huggingface-cli download Wan-AI/Wan2.2-I2V-A14B-Diffusers
Warning
Models will be downloaded to the container’s filesystem and will be lost when the container is removed unless you persist the data with a volume.
If you already have models downloaded on your host system, you can mount your existing cache.
Set your Hugging Face cache location.
export HF_HOME=/your/hf_cache/location
Download the model (if not already cached).
huggingface-cli download black-forest-labs/FLUX.1-dev
Launch the container with mounted cache.
docker run \ -it --rm \ --cap-add=SYS_PTRACE \ --security-opt seccomp=unconfined \ --user root \ --device=/dev/kfd \ --device=/dev/dri \ --group-add video \ --ipc=host \ --network host \ --privileged \ --shm-size 128G \ --name pytorch-xdit \ -e HSA_NO_SCRATCH_RECLAIM=1 \ -e OMP_NUM_THREADS=16 \ -e CUDA_VISIBLE_DEVICES=0,1,2,3,4,5,6,7 \ -e HF_HOME=/app/huggingface_models \ -v $HF_HOME:/app/huggingface_models \ rocm/pytorch-xdit:v26.1
If you prefer to keep the container self-contained or don’t have an existing cache.
Launch the container
docker run \ -it --rm \ --cap-add=SYS_PTRACE \ --security-opt seccomp=unconfined \ --user root \ --device=/dev/kfd \ --device=/dev/dri \ --group-add video \ --ipc=host \ --network host \ --privileged \ --shm-size 128G \ --name pytorch-xdit \ -e HSA_NO_SCRATCH_RECLAIM=1 \ -e OMP_NUM_THREADS=16 \ -e CUDA_VISIBLE_DEVICES=0,1,2,3,4,5,6,7 \ rocm/pytorch-xdit:v26.1
Inside the container, set the Hugging Face cache location and download the model.
export HF_HOME=/app/huggingface_models huggingface-cli download black-forest-labs/FLUX.1-dev
Warning
Models will be downloaded to the container’s filesystem and will be lost when the container is removed unless you persist the data with a volume.
If you already have models downloaded on your host system, you can mount your existing cache.
Set your Hugging Face cache location.
export HF_HOME=/your/hf_cache/location
Download the model (if not already cached).
huggingface-cli download black-forest-labs/FLUX.1-Kontext-dev
Launch the container with mounted cache.
docker run \ -it --rm \ --cap-add=SYS_PTRACE \ --security-opt seccomp=unconfined \ --user root \ --device=/dev/kfd \ --device=/dev/dri \ --group-add video \ --ipc=host \ --network host \ --privileged \ --shm-size 128G \ --name pytorch-xdit \ -e HSA_NO_SCRATCH_RECLAIM=1 \ -e OMP_NUM_THREADS=16 \ -e CUDA_VISIBLE_DEVICES=0,1,2,3,4,5,6,7 \ -e HF_HOME=/app/huggingface_models \ -v $HF_HOME:/app/huggingface_models \ rocm/pytorch-xdit:v26.1
If you prefer to keep the container self-contained or don’t have an existing cache.
Launch the container
docker run \ -it --rm \ --cap-add=SYS_PTRACE \ --security-opt seccomp=unconfined \ --user root \ --device=/dev/kfd \ --device=/dev/dri \ --group-add video \ --ipc=host \ --network host \ --privileged \ --shm-size 128G \ --name pytorch-xdit \ -e HSA_NO_SCRATCH_RECLAIM=1 \ -e OMP_NUM_THREADS=16 \ -e CUDA_VISIBLE_DEVICES=0,1,2,3,4,5,6,7 \ rocm/pytorch-xdit:v26.1
Inside the container, set the Hugging Face cache location and download the model.
export HF_HOME=/app/huggingface_models huggingface-cli download black-forest-labs/FLUX.1-Kontext-dev
Warning
Models will be downloaded to the container’s filesystem and will be lost when the container is removed unless you persist the data with a volume.
If you already have models downloaded on your host system, you can mount your existing cache.
Set your Hugging Face cache location.
export HF_HOME=/your/hf_cache/location
Download the model (if not already cached).
huggingface-cli download black-forest-labs/FLUX.2-dev
Launch the container with mounted cache.
docker run \ -it --rm \ --cap-add=SYS_PTRACE \ --security-opt seccomp=unconfined \ --user root \ --device=/dev/kfd \ --device=/dev/dri \ --group-add video \ --ipc=host \ --network host \ --privileged \ --shm-size 128G \ --name pytorch-xdit \ -e HSA_NO_SCRATCH_RECLAIM=1 \ -e OMP_NUM_THREADS=16 \ -e CUDA_VISIBLE_DEVICES=0,1,2,3,4,5,6,7 \ -e HF_HOME=/app/huggingface_models \ -v $HF_HOME:/app/huggingface_models \ rocm/pytorch-xdit:v26.1
If you prefer to keep the container self-contained or don’t have an existing cache.
Launch the container
docker run \ -it --rm \ --cap-add=SYS_PTRACE \ --security-opt seccomp=unconfined \ --user root \ --device=/dev/kfd \ --device=/dev/dri \ --group-add video \ --ipc=host \ --network host \ --privileged \ --shm-size 128G \ --name pytorch-xdit \ -e HSA_NO_SCRATCH_RECLAIM=1 \ -e OMP_NUM_THREADS=16 \ -e CUDA_VISIBLE_DEVICES=0,1,2,3,4,5,6,7 \ rocm/pytorch-xdit:v26.1
Inside the container, set the Hugging Face cache location and download the model.
export HF_HOME=/app/huggingface_models huggingface-cli download black-forest-labs/FLUX.2-dev
Warning
Models will be downloaded to the container’s filesystem and will be lost when the container is removed unless you persist the data with a volume.
If you already have models downloaded on your host system, you can mount your existing cache.
Set your Hugging Face cache location.
export HF_HOME=/your/hf_cache/location
Download the model (if not already cached).
huggingface-cli download stabilityai/stable-diffusion-3.5-large
Launch the container with mounted cache.
docker run \ -it --rm \ --cap-add=SYS_PTRACE \ --security-opt seccomp=unconfined \ --user root \ --device=/dev/kfd \ --device=/dev/dri \ --group-add video \ --ipc=host \ --network host \ --privileged \ --shm-size 128G \ --name pytorch-xdit \ -e HSA_NO_SCRATCH_RECLAIM=1 \ -e OMP_NUM_THREADS=16 \ -e CUDA_VISIBLE_DEVICES=0,1,2,3,4,5,6,7 \ -e HF_HOME=/app/huggingface_models \ -v $HF_HOME:/app/huggingface_models \ rocm/pytorch-xdit:v26.1
If you prefer to keep the container self-contained or don’t have an existing cache.
Launch the container
docker run \ -it --rm \ --cap-add=SYS_PTRACE \ --security-opt seccomp=unconfined \ --user root \ --device=/dev/kfd \ --device=/dev/dri \ --group-add video \ --ipc=host \ --network host \ --privileged \ --shm-size 128G \ --name pytorch-xdit \ -e HSA_NO_SCRATCH_RECLAIM=1 \ -e OMP_NUM_THREADS=16 \ -e CUDA_VISIBLE_DEVICES=0,1,2,3,4,5,6,7 \ rocm/pytorch-xdit:v26.1
Inside the container, set the Hugging Face cache location and download the model.
export HF_HOME=/app/huggingface_models huggingface-cli download stabilityai/stable-diffusion-3.5-large
Warning
Models will be downloaded to the container’s filesystem and will be lost when the container is removed unless you persist the data with a volume.
If you already have models downloaded on your host system, you can mount your existing cache.
Set your Hugging Face cache location.
export HF_HOME=/your/hf_cache/location
Download the model (if not already cached).
huggingface-cli download Tongyi-MAI/Z-Image-Turbo
Launch the container with mounted cache.
docker run \ -it --rm \ --cap-add=SYS_PTRACE \ --security-opt seccomp=unconfined \ --user root \ --device=/dev/kfd \ --device=/dev/dri \ --group-add video \ --ipc=host \ --network host \ --privileged \ --shm-size 128G \ --name pytorch-xdit \ -e HSA_NO_SCRATCH_RECLAIM=1 \ -e OMP_NUM_THREADS=16 \ -e CUDA_VISIBLE_DEVICES=0,1,2,3,4,5,6,7 \ -e HF_HOME=/app/huggingface_models \ -v $HF_HOME:/app/huggingface_models \ rocm/pytorch-xdit:v26.1
If you prefer to keep the container self-contained or don’t have an existing cache.
Launch the container
docker run \ -it --rm \ --cap-add=SYS_PTRACE \ --security-opt seccomp=unconfined \ --user root \ --device=/dev/kfd \ --device=/dev/dri \ --group-add video \ --ipc=host \ --network host \ --privileged \ --shm-size 128G \ --name pytorch-xdit \ -e HSA_NO_SCRATCH_RECLAIM=1 \ -e OMP_NUM_THREADS=16 \ -e CUDA_VISIBLE_DEVICES=0,1,2,3,4,5,6,7 \ rocm/pytorch-xdit:v26.1
Inside the container, set the Hugging Face cache location and download the model.
export HF_HOME=/app/huggingface_models huggingface-cli download Tongyi-MAI/Z-Image-Turbo
Warning
Models will be downloaded to the container’s filesystem and will be lost when the container is removed unless you persist the data with a volume.
Run inference#
Clone the ROCm Model Automation and Dashboarding (ROCm/MAD) repository to a local directory and install the required packages on the host machine.
git clone https://github.com/ROCm/MAD cd MAD pip install -r requirements.txt
On the host machine, use this command to run the performance benchmark test on the Hunyuan Video model using one node.
export MAD_SECRETS_HFTOKEN="your personal Hugging Face token to access gated models" madengine run \ --tags pyt_xdit_hunyuanvideo \ --keep-model-dir \ --live-output
MAD launches a Docker container with the name
container_ci-pyt_xdit_hunyuanvideo. The throughput and serving reports of the
model are collected in the following paths: pyt_xdit_hunyuanvideo_throughput.csv
and pyt_xdit_hunyuanvideo_serving.csv.
To run the benchmarks for Hunyuan Video, use the following command:
cd /app/Hunyuanvideo
mkdir results
torchrun --nproc_per_node=8 run.py \
--model tencent/HunyuanVideo \
--prompt "In the large cage, two puppies were wagging their tails at each other." \
--batch_size 1 \
--height 720 --width 1280 \
--seed 1168860793 \
--num_frames 129 \
--num_inference_steps 50 \
--warmup_steps 1 \
--n_repeats 1 \
--sleep_dur 10 \
--ulysses_degree 8 \
--enable_tiling --enable_slicing \
--guidance_scale 6.0 \
--use_torch_compile \
--attention_backend aiter \
--benchmark_output_directory results
The generated video will be stored under the results directory.
Clone the ROCm Model Automation and Dashboarding (ROCm/MAD) repository to a local directory and install the required packages on the host machine.
git clone https://github.com/ROCm/MAD cd MAD pip install -r requirements.txt
On the host machine, use this command to run the performance benchmark test on the Hunyuan Video 1.5 model using one node.
export MAD_SECRETS_HFTOKEN="your personal Hugging Face token to access gated models" madengine run \ --tags pyt_xdit_hunyuanvideo_1_5 \ --keep-model-dir \ --live-output
MAD launches a Docker container with the name
container_ci-pyt_xdit_hunyuanvideo_1_5. The throughput and serving reports of the
model are collected in the following paths: pyt_xdit_hunyuanvideo_1_5_throughput.csv
and pyt_xdit_hunyuanvideo_1_5_serving.csv.
To run the benchmarks for Hunyuan Video 1.5, use the following command:
cd /app/Hunyuanvideo_1_5
mkdir results
torchrun --nproc_per_node=8 /app/Hunyuanvideo_1_5/run.py \
--model hunyuanvideo-community/HunyuanVideo-1.5-Diffusers-720p_t2v \
--prompt "In the large cage, two puppies were wagging their tails at each other." \
--task t2v \
--height 720 --width 1280 \
--seed 1168860793 \
--num_frames 129 \
--num_inference_steps 50 \
--num_repetitions 1 \
--ulysses_degree 8 \
--enable_tiling --enable_slicing \
--use_torch_compile \
--attention_backend aiter \
--benchmark_output_directory results
The generated video will be stored under the results directory.
Clone the ROCm Model Automation and Dashboarding (ROCm/MAD) repository to a local directory and install the required packages on the host machine.
git clone https://github.com/ROCm/MAD cd MAD pip install -r requirements.txt
On the host machine, use this command to run the performance benchmark test on the Wan2.1 model using one node.
export MAD_SECRETS_HFTOKEN="your personal Hugging Face token to access gated models" madengine run \ --tags pyt_xdit_wan_2_1 \ --keep-model-dir \ --live-output
MAD launches a Docker container with the name
container_ci-pyt_xdit_wan_2_1. The throughput and serving reports of the
model are collected in the following paths: pyt_xdit_wan_2_1_throughput.csv
and pyt_xdit_wan_2_1_serving.csv.
To run the benchmarks for Wan2.1, use the following command:
cd /app/Wan
mkdir results
torchrun --nproc_per_node=8 /app/Wan/run.py \
--model Wan-AI/Wan2.1-I2V-14B-720P-Diffusers \
--prompt "Summer beach vacation style, a white cat wearing sunglasses sits on a surfboard. The fluffy-furred feline gazes directly at the camera with a relaxed expression. Blurred beach scenery forms the background featuring crystal-clear waters, distant green hills, and a blue sky dotted with white clouds. The cat assumes a naturally relaxed posture, as if savoring the sea breeze and warm sunlight. A close-up shot highlights the feline's intricate details and the refreshing atmosphere of the seaside." \
--task i2v \
--height 720 \
--width 1280 \
--img_file_path /app/Wan/i2v_input.JPG \
--num_frames 81 \
--ulysses_degree 8 \
--seed 42 \
--num_repetitions 1 \
--num_inference_steps 40 \
--use_torch_compile \
--attention_backend aiter \
--benchmark_output_directory results
The generated video will be stored under the results directory.
Clone the ROCm Model Automation and Dashboarding (ROCm/MAD) repository to a local directory and install the required packages on the host machine.
git clone https://github.com/ROCm/MAD cd MAD pip install -r requirements.txt
On the host machine, use this command to run the performance benchmark test on the Wan2.2 model using one node.
export MAD_SECRETS_HFTOKEN="your personal Hugging Face token to access gated models" madengine run \ --tags pyt_xdit_wan_2_2 \ --keep-model-dir \ --live-output
MAD launches a Docker container with the name
container_ci-pyt_xdit_wan_2_2. The throughput and serving reports of the
model are collected in the following paths: pyt_xdit_wan_2_2_throughput.csv
and pyt_xdit_wan_2_2_serving.csv.
To run the benchmarks for Wan2.2, use the following command:
cd /app/Wan
mkdir results
torchrun --nproc_per_node=8 /app/Wan/run.py \
--model Wan-AI/Wan2.2-I2V-A14B-Diffusers \
--prompt "Summer beach vacation style, a white cat wearing sunglasses sits on a surfboard. The fluffy-furred feline gazes directly at the camera with a relaxed expression. Blurred beach scenery forms the background featuring crystal-clear waters, distant green hills, and a blue sky dotted with white clouds. The cat assumes a naturally relaxed posture, as if savoring the sea breeze and warm sunlight. A close-up shot highlights the feline's intricate details and the refreshing atmosphere of the seaside." \
--task i2v \
--height 720 \
--width 1280 \
--img_file_path /app/Wan/i2v_input.JPG \
--num_frames 81 \
--ulysses_degree 8 \
--seed 42 \
--num_repetitions 1 \
--num_inference_steps 40 \
--use_torch_compile \
--attention_backend aiter \
--benchmark_output_directory results
The generated video will be stored under the results directory.
Clone the ROCm Model Automation and Dashboarding (ROCm/MAD) repository to a local directory and install the required packages on the host machine.
git clone https://github.com/ROCm/MAD cd MAD pip install -r requirements.txt
On the host machine, use this command to run the performance benchmark test on the FLUX.1 model using one node.
export MAD_SECRETS_HFTOKEN="your personal Hugging Face token to access gated models" madengine run \ --tags pyt_xdit_flux \ --keep-model-dir \ --live-output
MAD launches a Docker container with the name
container_ci-pyt_xdit_flux. The throughput and serving reports of the
model are collected in the following paths: pyt_xdit_flux_throughput.csv
and pyt_xdit_flux_serving.csv.
To run the benchmarks for FLUX.1, use the following command:
cd /app/Flux
mkdir results
torchrun --nproc_per_node=8 /app/Flux/run.py \
--model black-forest-labs/FLUX.1-dev \
--seed 42 \
--prompt "A small cat" \
--height 1024 \
--width 1024 \
--num_inference_steps 25 \
--max_sequence_length 256 \
--warmup_steps 5 \
--no_use_resolution_binning \
--ulysses_degree 8 \
--use_torch_compile \
--guidance_scale 0.0 \
--num_repetitions 50 \
--attention_backend aiter \
--benchmark_output_directory results
The generated video will be stored under the results directory.
You may also use run_usp.py which implements USP without modifying the default diffusers pipeline.
Clone the ROCm Model Automation and Dashboarding (ROCm/MAD) repository to a local directory and install the required packages on the host machine.
git clone https://github.com/ROCm/MAD cd MAD pip install -r requirements.txt
On the host machine, use this command to run the performance benchmark test on the FLUX.1 Kontext model using one node.
export MAD_SECRETS_HFTOKEN="your personal Hugging Face token to access gated models" madengine run \ --tags pyt_xdit_flux_kontext \ --keep-model-dir \ --live-output
MAD launches a Docker container with the name
container_ci-pyt_xdit_flux_kontext. The throughput and serving reports of the
model are collected in the following paths: pyt_xdit_flux_kontext_throughput.csv
and pyt_xdit_flux_kontext_serving.csv.
To run the benchmarks for FLUX.1 Kontext, use the following command:
cd /app/Flux
mkdir results
torchrun --nproc_per_node=8 /app/Flux/run_usp.py \
--model black-forest-labs/FLUX.1-Kontext-dev \
--seed 42 \
--prompt "Add a cool hat to the cat" \
--height 1024 \
--width 1024 \
--num_inference_steps 30 \
--max_sequence_length 512 \
--warmup_steps 5 \
--no_use_resolution_binning \
--ulysses_degree 8 \
--use_torch_compile \
--img_file_path /app/Flux/cat.png \
--model_type flux_kontext \
--guidance_scale 2.5 \
--num_repetitions 25 \
--attention_backend aiter \
--benchmark_output_directory results
The generated video will be stored under the results directory.
Clone the ROCm Model Automation and Dashboarding (ROCm/MAD) repository to a local directory and install the required packages on the host machine.
git clone https://github.com/ROCm/MAD cd MAD pip install -r requirements.txt
On the host machine, use this command to run the performance benchmark test on the FLUX.2 model using one node.
export MAD_SECRETS_HFTOKEN="your personal Hugging Face token to access gated models" madengine run \ --tags pyt_xdit_flux_2 \ --keep-model-dir \ --live-output
MAD launches a Docker container with the name
container_ci-pyt_xdit_flux_2. The throughput and serving reports of the
model are collected in the following paths: pyt_xdit_flux_2_throughput.csv
and pyt_xdit_flux_2_serving.csv.
To run the benchmarks for FLUX.2, use the following command:
cd /app/Flux
mkdir results
torchrun --nproc_per_node=8 /app/Flux/run_usp.py \
--model black-forest-labs/FLUX.2-dev \
--seed 42 \
--prompt "Add a cool hat to the cat" \
--height 1024 \
--width 1024 \
--num_inference_steps 50 \
--max_sequence_length 512 \
--warmup_steps 5 \
--no_use_resolution_binning \
--ulysses_degree 8 \
--use_torch_compile \
--img_file_paths /app/Flux/cat.png \
--model_type flux2 \
--guidance_scale 4.0 \
--num_repetitions 25 \
--attention_backend aiter \
--benchmark_output_directory results
The generated video will be stored under the results directory.
Clone the ROCm Model Automation and Dashboarding (ROCm/MAD) repository to a local directory and install the required packages on the host machine.
git clone https://github.com/ROCm/MAD cd MAD pip install -r requirements.txt
On the host machine, use this command to run the performance benchmark test on the stable-diffusion-3.5-large model using one node.
export MAD_SECRETS_HFTOKEN="your personal Hugging Face token to access gated models" madengine run \ --tags pyt_xdit_sd_3_5 \ --keep-model-dir \ --live-output
MAD launches a Docker container with the name
container_ci-pyt_xdit_sd_3_5. The throughput and serving reports of the
model are collected in the following paths: pyt_xdit_sd_3_5_throughput.csv
and pyt_xdit_sd_3_5_serving.csv.
To run the benchmarks for stable-diffusion-3.5-large, use the following command:
cd /app/StableDiffusion3.5
mkdir results
torchrun --nproc_per_node=8 /app/StableDiffusion3.5/run.py \
--model stabilityai/stable-diffusion-3.5-large \
--prompt "A capybara holding a sign that reads Hello World" \
--num_repetitions 50 \
--num_inference_steps 28 \
--pipefusion_parallel_degree 4 \
--use_cfg_parallel \
--use_torch_compile \
--dtype torch.float16 \
--attention_backend aiter \
--benchmark_output_directory results
The generated video will be stored under the results directory.
Clone the ROCm Model Automation and Dashboarding (ROCm/MAD) repository to a local directory and install the required packages on the host machine.
git clone https://github.com/ROCm/MAD cd MAD pip install -r requirements.txt
On the host machine, use this command to run the performance benchmark test on the Z-Image Turbo model using one node.
export MAD_SECRETS_HFTOKEN="your personal Hugging Face token to access gated models" madengine run \ --tags pyt_xdit_z_image_turbo \ --keep-model-dir \ --live-output
MAD launches a Docker container with the name
container_ci-pyt_xdit_z_image_turbo. The throughput and serving reports of the
model are collected in the following paths: pyt_xdit_z_image_turbo_throughput.csv
and pyt_xdit_z_image_turbo_serving.csv.
To run the benchmarks for Z-Image Turbo, use the following command:
cd /app/Z-Image
mkdir results
torchrun --nproc_per_node=2 /app/Z-Image/run.py \
--model Tongyi-MAI/Z-Image-Turbo \
--seed 42 \
--prompt "A crowded beach" \
--height 1088 \
--width 1920 \
--num_inference_steps 9 \
--ulysses_degree 2 \
--use_torch_compile \
--guidance_scale 0.0 \
--num_repetitions 50 \
--attention_backend aiter \
--benchmark_output_directory results
The generated video will be stored under the results directory.