6,000 Episodes and a GPU Bill: Inside My Weird Prompts

A sloth, a donkey, and a curious human send a prompt and see what they're doing. One hits a record with a question on his mind. Two brothers pick it up, see what they find. Corn takes his time. Herman's got the facts. Back and forth, they never hold back. My Weird Prompts, come on in. That's where we begin. My Weird Prompts, glad you're here.

Hey, everyone. Corn here, and Herman's with me, but not for long because today we're handing the microphone over.

That's right. This one is a creator update. Daniel, the human who sends us every one of these weird prompts, recorded some notes on where the show has been, where he wants to take it, and what's been changing in the pipeline behind the scenes.

So no deep dive from us this time, just Daniel in his own voice telling you what's going on.

Which, honestly, is the only way I ever find out what he's planning for us.

Fair. Over to you, Daniel.

Well, hello there to anyone who listens to the My Weird Prompts podcast. My name is Daniel Rozell. I live in Jerusalem, and I am the human in the loop, the permanent human in the loop, uh, behind the My Weird Prompts podcast. And we're at almost 6,000 episodes. In fact, I think that as I speak, uh, we're gonna publish our 6,000th episode. And I have to say, I feel kind of slightly a strange mixture of ashamed, apologetic, and, um, maybe just those two things about it, as well as having quite a large bill to pay to Modal. I don't want to see this month's, um, GPU bill, uh, for Modal serverless inference. Modal has been an absolutely Trojan, um, platform for the podcast since its very inception. Um, I did look at the start at running Chatterbox locally, and on my flimsy eight gigabyte VRAM GPU, I think Claude Code said, "Yeah, it'll take you about 20..." and then it kind of hung, uh, hours to generate an episode. I was like, "Yeah, that's a bit slow." Um, we did manage to get our generation time per episode down a fraction a few days ago. Um, some changes in decoding. And what I wanted to do in this episode was, it's been such a while since I actually spoke myself on the podcast, and I, I think I've done one of these before. It was, uh, at year-end, I believe, uh, which is amazing to think. This has been just the quickest year, uh, of my life. Feels like the quickest year. That feels like just a few weeks ago that I, uh, was wrapping up 2025. And so it's been almost a year, uh, since I did that, which must mean, I guess, the podcast has been running for that long, and I never intended it to be this, uh, to have grown to such a huge amount of episodes. And the fact that I've been able to do this much without going totally broke, um, is a testament to, well, I would say imaginative engineering, Claude Code being just sensational, um, at figuring out all the bits and pieces. I kind of bring the ideas occasionally, um, look at the pipeline code, which is, at this point, um, a one main, one main Python process, and then a lot of nested nodes. We're moving much more towards a very modularized, uh, production pipeline. But for the most part, I just sit there holding down the space bar and, uh, saying, "Yeah, this isn't really working well. Um, what can we-- How can we do