Submissions by s-macke

The simplest, fastest repository for training/finetuning small-sized VLMs

Gemini 2.5 Pro won Pokémon Blue in 106k moves

SWE-Smith: Scaling Data for Software Engineering Agents

Sam Altman: we added one million users in the last hour

Measuring AI Ability to Complete Long Tasks

FrontierMath Was Funded by OpenAI

Grokking at the Edge of Numerical Stability

LLMs struggle with perception, not reasoning, in ARC-AGI

A Llama 70B finetune that has reflection baked into it's weights