Subdomain

What experts are saying about LLMs

Large language models — the architecture, training, capabilities, and limits of the models powering modern AI.

1,615 claims tagged

Episodes discussing LLMs

The math behind how LLMs are trained and served – Reiner Pope

The RL Fine-Tuning Playbook: CoreWeave's Kyle Corbitt on GRPO, Rubrics, Environments, Reward Hacking

Is It Time to Rethink LLM Pre-Training? [Aditi Raghunathan] - 747

Amber Teng - Building apps with a new generation of language models

Elon’s Anthropic Deal, The Next AI Monopoly?, “FDA for AI” Panic, Trading the AI Boom

The history and future of AI at Google, with Sundar Pichai

Google's Liz Reid on Who Will Own Search in a World of AI | Odd Lots

Inside a16z’s Top 100 AI Apps Report with Olivia Moore

People discussing LLMs

Dylan Patel22 Sundar Pichai21 Gavin Baker20 Mandeep Singh19 Nathan Labenz18 Edwin Chen16 Jason Calacanis15 Ed Ludlow14

Sample claims

“Plaid's CEO Zach Perret uses custom-built GPTs for various tasks, including one he created to act as his own executive coach by feeding it text from books and transcripts.”

— Zach Perret

“Google's advanced reasoning model, Gemini 3.1 Pro, is now available in preview and is optimized for complex workflow orchestration.”

— Thomas Kurian

“Google's Gemini 3 foundation model runs on the company's in-house designed and built infrastructure.”

— Amin Vahdat

“Google announced Gemini 3.1 Flash Image (Nano Banana 2), Deo 3-1 Lite, and Lyria 3 Pro models, all available in preview.”

— Thomas Kurian

“Google's PaLM model had 540 billion parameters, and after its release, companies generally stopped publishing parameter counts for their large language models.”

— Akansha Chowdhury