.jpeg)
Multi-Instance GPUs on Runpod: Stop Paying for Compute You Don't Need
With MIG, we can partition RTX 6000 Pro cards into isolated 24 GB instances. Here's when it makes sense for your workloads.
Blog
Curious about JAX-based diffusion models? This post walks through setting up and running them on Runpod using our GPU pods. It covers environment setup, model launching, and highlights the performance benefits of JAX for image generation workflows.

Here's a quick and dirty guide to getting Huemin's JAX diffusion to run on runpod!
Today I figured I'd switch things up a bit and leave you a gif quick start instead of the usual article. It only takes a few minutes to get started with this and all the deps have been sorted out for you in the nb. If you need the notebook file that you should upload to your instance, you can download it from one of these two places: Link 1, Link 2

Please note that it will sit here for quite a few minutes while it downloads models and prepared for the first render cycle:

Be patient and it will start rendering when it's ready. I was getting around 1.5-2s/it on the demo example.
Enjoy!
Credits to: Huemin (@huemin_art), for the notebook, Alexander Redde (@alexanderredde3) for working out the deps, and nsheppherd (@nshepperd1) and Rivers Have Wings (@rivershavewings) for Jax Diffusion
Blog Posts
.jpeg)
With MIG, we can partition RTX 6000 Pro cards into isolated 24 GB instances. Here's when it makes sense for your workloads.

How 1,100 researchers beat OpenAI's own baseline with 16 megabytes and 10 minutes.

Learn how to set up a real-world agentic system with our new Flash framework.