LaIA
A web-search and video creation assistant designed to make public information more accessible to young people.
Project ledger
Projects
Open models, datasets, experiments, and tools built to understand the full machine-learning stack.
→
Cucafera
244M parameters, 5.5B training tokens, one A100 and a transparent record of how it was made.
Patufet
Pretraining text, educational material, stories, code, prompts, conversations, and instruction data.
This is a selection, where each project is included for the idea it explores. It explains the decisions and processes.