Live data from Hacker News

Nanocode: The best Claude Code that $200 can buy in pure JAX on TPUs

github.com

31–33 of 33 posts

Re: Nanocode: The best Claude Code that $200 can buy in pure JAX on TPUs

#32

[flagged]

Yes my findings and thoughts were pretty much identical. I actually think you can get something reasonable at 1.3B params with the correct training recipe, but definitely not at this compute/token budget.

One thing I found was that the model would pretty much always emit solutions from its training data when asked to solve problems, but it was much better at using Bash commands to explore a codebase, for example.

The Hugging Face folks have a great post on also using CAI for more vibes/character post-training than harmlessness https://huggingface.co/blog/constitutional_ai#oh-honey-lets-...

Re: Nanocode: The best Claude Code that $200 can buy in pure JAX on TPUs

#33
post #28

Earlier quoted context omitted.

your second function is the type of bad code you get from people trying to program python like its c

Is there a pythonic way to satisfy the prompt? IE without making a new list?

https://news.ycombinator.com/item?id=47652757
Post reply on HN