Claude 3.5 Haiku

LLMs and Newcomb's Problem - Computerphile

Computerphile

Model predicted to act in a certain way (one box) in the demonstration.

The new Claude 3.5 Sonnet, Computer Use, and Building SOTA Agents — with Erik Schluntz, Anthropic

Latent Space

Anthropic's smallest and fastest model, which still performed surprisingly well on the Sweet-Bench benchmark.

Gemini 2.5 Pro - It’s a Darn Smart Chatbot … (New Simple High Score)

AI Explained

The Utility of Interpretability — Emmanuel Amiesen

Latent Space

A specific version of Anthropic's Claude model, which displays planning capabilities in poetry generation, able to backward-plan its sentences based on a desired rhyme at the end.