Untitled entry

Summary from the TLDR AI newsletter

It doesn't really make sense to run a massive inference model for every small task. That's why Fastino is introducing TLMs [4] - task-specific Language Models that are better, faster, and cheaper.