Skip to content


Folders and files

Last commit message
Last commit date

Latest commit



2 Commits

Repository files navigation

Denoising Diffusion Probabilistic Models

Jonathan Ho, Ajay Jain, Pieter Abbeel



Samples generated by our model

Experiments run on Google Cloud TPU v3-8. Requires TensorFlow 1.15 and Python 3.5, and these dependencies for CPU instances (see requirements.txt):

pip3 install fire
pip3 install scipy
pip3 install pillow
pip3 install tensorflow-probability==0.8
pip3 install tensorflow-gan==0.0.0.dev0
pip3 install tensorflow-datasets==2.1.0

The training and evaluation scripts are in the scripts/ subdirectory. The commands to run training and evaluation are in comments at the top of the scripts. Data is stored in GCS buckets. The scripts are written to assume that the bucket names are of the form gs://mybucketprefix-us-central1; i.e. some prefix followed by the region. The prefix should be passed into the scripts using the --bucket_name_prefix flag.

Models and samples can be found at:


If you find our work relevant to your research, please cite:

    title={Denoising Diffusion Probabilistic Models},
    author={Jonathan Ho and Ajay Jain and Pieter Abbeel},
    journal={arXiv preprint arxiv:2006.11239}