
03: The Next Generation of LLMs with Jonathan Frankle of MosaicML
Replit AI Podcast
The 3B Model: A Sweet Spot for Software Development
The 3B model is portable enough to run on a computer from 10 years ago. It's really fun to see just the community hacking and playing with it, so that was a consideration for us. I'm kind of optimistic that without too much trouble, we may be able to get a model that is weird shape for H 100s and is easily adaptable for CPUs. There's so much more we can do. And I'm really excited about what our next steps are.
00:00
Transcript
Play full episode
Remember Everything You Learn from Podcasts
Save insights instantly, chat with episodes, and build lasting knowledge - all powered by AI.