Controlling Stable Diffusion with TPUs ⚡️ 
Google partners up with Hugging Face to control Stable Diffusion with free TPUs! 

ControlNet is a neural network structure that allows fine-grained control of diffusion models by adding extra conditions. In this event, we will be applying ControlNet and various other fine-tuning techniques on Stable Diffusion with JAX and Diffusers, using free TPU hours from Google Cloud. 

To see what ControlNet is capable of, you can see this blog post and check out this example application that includes various controls.  

To get started, simply fill out this form and join the discord (more instructions below).

Email *
Please provide a gmail here for TPU access.
May we ask you your first name? *
Your Hugging Face Hub username? *
We will be having this community event on Hugging Face Discord, you can simply join here 👉 huggingface.co/join/discord
After you join the discord, take the role Diffusers from #role-assignment and head to #jax-diffusers-ideas forum channel to share your project idea and find teammates. 

After this form is submitted, we will be evaluating the responses and granting participants with TPUs and giving them access to sprint-related channels. Make sure to keep an eye on your e-mail inbox!
Your Discord username? (Please write full username with ID, e.g. huggingface#1234) *
Glad to see you here! Would you rather describe yourself as "experienced" or "newbie" in machine learning and diffusion models?
 *
(Optional) Explain your prior experience (if exists) with following:
- machine learning,
- diffusion models,
- JAX/Flax,
- ControlNet
Do you already have a cool idea what you would like to build in this event? If so, can you explain? If not, that's totally fine as well - there are many other teams/projects you can join! In this case simply reply with "Not yet". *
(Mandatory) Please read Diffusers' ethical charter and Google AI Principles and agree to the conditions by checking the box below.  *
Required
Submit
Clear form
Never submit passwords through Google Forms.
reCAPTCHA
PrivacyTerms
This form was created inside of Hugging Face. Report Abuse
Google Forms