3min snip

Gradient Dissent: Conversations on AI cover image

Emad Mostaque — Stable Diffusion, Stability AI, and What’s Next

Gradient Dissent: Conversations on AI

NOTE

**Discussion about model size **

The speaker predicts a reverse of the current trend of building bigger models, suggesting that smaller models might become more prevalent due to the unnecessary size of current models. They question the necessity of large model sizes and suggest that smaller, more optimized models with a focused diet of training data may be more effective. They also raise the question of why text models are so much larger than image models and highlight that the amount of data and optimization play crucial roles in model performance, showing that training and data optimization can lead to better performance than simply increasing model size.

00:00

Get the Snipd
podcast app

Unlock the knowledge in podcasts with the podcast player of the future.
App store bannerPlay store banner

AI-powered
podcast player

Listen to all your favourite podcasts with AI-powered features

Discover
highlights

Listen to the best highlights from the podcasts you love and dive into the full episode

Save any
moment

Hear something you like? Tap your headphones to save it with AI-generated key takeaways

Share
& Export

Send highlights to Twitter, WhatsApp or export them to Notion, Readwise & more

AI-powered
podcast player

Listen to all your favourite podcasts with AI-powered features

Discover
highlights

Listen to the best highlights from the podcasts you love and dive into the full episode