RunPod Blog

Here's a quick and dirty guide to getting Huemin's JAX diffusion to run on runpod!

Image generation has evolved tremendously over the last couple of years. Generative Adversarial Networks have emerged as state-of-the-art in the field of image generation. The latest of which is StyleGAN3[1].

You could schedule advanced stuff like cuts before, but now you can schedule PROMPTS in Disco Diffusion! That means that you can alter the prompt over the course of your diffusion steps to achieve some really awesome results. Read more to find out how!

Have you ever wanted to get your hands dirty with Disco Diffusion, but aren't a big fan of making sure your syntax is right? What about if you wanted to do a bunch of different settings variations in one go? While there are plenty of options for generative art with DD out there, none of them are both user friendly AND give you the power to generate a large volume with advanced settings. The wait is over, version 1 of RunPod's user interface is here!

We've been sailing smoothly with entmike's fork of Disco Diffusion for a little while now. We recently caught wind of the discoart fork and wow is it good. We've made the decision to migrate both RunPod and Fever Dreams to use discoart as a base. Shoutout to Jina.ai for their wonderful work!

If you didn't know, every Disco Diffusion render needs to start with some kind of input. If you aren't using an init image, this input is perlin noise, which is kind of like old school TV static. In this article, we're going to introduce another method of initialing your render: voronoi. You can get some pretty interesting results with this, so check it out!

While the entmike fork on RunPod is pretty rad, we do get that some of you may want to try the latest and greatest features and can't wait for the 10 seconds it takes Mike to catch up ;). While running vanilla DD on RunPod has always been an option, we've decided to make it even easier to get started by integrating the notebook into our default run template.

Google Colab Pro Plus costs $50 with a chance to get a V100 or (in rare cases) a A100 GPU. While, with RunPod's GPU Cloud pay-as-you go model, you can get guaranteed GPU compute for as low as $0.2/hour.

At RunPod, we're always looking to offer more value to our clients. One of the things we take seriously is security. To this end, we have implemented the ability to encrypt your persistent volumes.

Have you ever really wanted to generate some art using disco diffusion, but you aren't sure what to put in? I think we've all been there! Thankfully, there's a "prompt salad" feature in our version of disco diffusion that can help with your artists block!

Have you seen those cool prompt/settings studies that people have done? Maybe you wanted to see what the different effects of cut_ic_power are on your final product, or play around with clip_guidance_scale, or to simply try a bunch of text prompt variations. You can automate generation of these exploratory studies using config files. Read on to find out how!

It's pretty easy to just click in some settings when you run Disco Diffusion for the first time and get some generally good results. However, if you really want to up your game, then using consistent settings each time and varying them slowly until you get the results you want is key. In this article, we'll go over how to get consistent results and save your progress so you waste less time clicking.

There are a few use cases for running a GPU accelerated virtual desktop on RunPod. One common use case is for those who need to use demanding applications that require a lot of processing power, such as 3D rendering with blender. In this blog post, we'll go over how to easily deploy a pod with everything you need to get rendering.

It looks like some of you are used to Google Colab's interface and would prefer to use that over the command line or Jupyterlab's interface. Well, good news, you can connect Colab to use your own custom compute instead of the GPU that they assign to you! Unfortunately, Google won't let you connect directly to a remote instance, but with a few simple steps, you can get around it by using your local machine to forward the remote port. Sound complicated? It's not too bad! We can get this going together in a few minutes :)

We've gotten a lot of questions around "spot" instances vs "on-demand" instances. While these terms are familiar to those who have used AWS EC2 instances in the past, it's clear that many people are not familiar with these terms. Let's do a quick summary of the pros and cons of each type of instance and go through examples of use cases for each instance type.

While RunPod offers persistent volume storage for your pods, we understand that many users would like to have a cloud backup in order to avoid idle storage charges or to be able to move their work to a new pod if the need arises. RunPod Cloud Sync offers several integrations with cloud storage providers like AWS S3, Google Cloud Buckets, and Azure Blob Storage. In this blog post, I'll go through the simple steps to achieve a cloud backup using the Backblaze B2 cloud storage service, as it is very affordable and easy to use.

While working in a terminal window is sometimes good enough, it's much nicer to work within your IDE if you are doing interactive development. Fortunately, you can connect your VSCode IDE to your RunPod instance via SSH. This will allow you to edit files and code on your remote instance directly from VSCode.

AI is the latest frontier for artists, providing new opportunities for creativity and expression. Disco Diffusion models are a great way to create some stunning pieces. In this article, we'll go through how to get running Disco Diffusion on RunPod.

Hi everyone, this isn't really an important event or anything, but you may have noticed that the RunPod blog has moved from blog.runpod.io to runpod.io/blog. At first glance, this seems like a pretty trivial change - I mean, why bother? It hasn't been a trivial transition, but hopefully it will be worth it in the end. Let's talk about why we ultimately decided to do this and what the pros and cons are (that we've found so far). Spoiler: worse editor, more customizability, better??? SEO.