I recently finetuned an LLM (gemma-3-270M) to extract recipes from blog posts.

My thoughts are that this is a perfect model for running locally on an iPhone in a recipe-management app.

Wrote a blog about what I learned while finetuning it https://vladr.com/blog/posts/finetuning-gemma3-to-extract-re...

And published the model on HF https://huggingface.co/v-rusu/recipe-extractor

Cool animations!

This was definitely one of the most disturbing experiences I've had.

But it's somehow awesome at the same time.

do you know if someone actually compared a 4o CoT to the o1? I'm trying to find something on it, but I can't find anything.

LE: I found this tweet by Catena Labs of their MoA mix compared to o1-preview: https://x.com/catena_labs/status/1834416060071571836