Sign in Subscribe

aimodels-fyi

SmolDocling: An Ultra-Compact VLM for Document Understanding

SmolDocling: An Ultra-Compact VLM for Document Understanding

Example of how rStar-Math works

Creating artificial doubt significantly improves AI math accuracy

Example generation.

Step-by-step reasoning can fix madman logic in vision AI

All LLMs use tokenization. Are we doing it totally wrong?

All LLMs use tokenization. Are we doing it totally wrong?

Slashing model size by 85% while redefining how we build adaptable, efficient LLMs

Imagen 3 generation results

Can image models understand what we’re asking for?

High-quality graphics vs high-quality understanding — which one matters more?

Example DreamRunner video generation, from the project website.

Teaching AI to tell visually consistent stories

Results of the BLT architecture, from the Meta paper

Bye tokens, hello patches

Meta announces a better way to scale LLMs