upvote

points

by iteria7 days ago |
comments
upvoteby simonw7 days ago|
[-]
When is the last time you tried using LLMs against a large, old, crufty undocumented codebase?

Things have changed a lot in the past six weeks.

Gemini 2.5 Pro accepts a million tokens and can "reason" with them, which means you can feed it hundreds of thousands of lines of code and it has a surprisingly good chance of figuring things out.

OpenAI released their first million token models with the GPT 4.1 series.

OpenAI o3 and o4-mini are both very strong reasoning code models with 200,000 token input limits.

These models are all new within the last six weeks. They're very, very good at working with large amounts of crufty undocumented code.

reply
upvoteby grugagag6 days ago|
parent|
next
[-]
Ultimately LLMs don’t really understand what the code does at runtime. Sure, just parsing out the codebase can help make a good guess but in some cases it’s hard to trust LLMs with changes because the consequences are unknown in complex codebases that have weird warts nobody documented.

Maybe in a generation or two codebases will become more uniform and predictible if fewer humans do it by hand. Same with self driving cars, if there were no human drivers out there the problem would become trivial to conquer.

reply
upvoteby simonw6 days ago|
parent|
next
[-]
That's a lot less true today than it was six weeks ago. The "reasoning" models are spookily good at answering questions about how code runs, and identifying the source of bugs.

They still make mistakes, and yeah they're still (mostly) next token predicting machines under the hood, but if your mental model is "they can't actually predict through how some code will execute" you may need to update that.

reply
upvoteby 5 days ago|
parent|
prev|
[-]
deleted
reply
upvoteby LunaSea6 days ago|
parent|
prev|
[-]
Gemini 2.5 Pro crashes with a 50) status code every 5 requests. Not great for a model you're supposed to rely on.
reply
upvoteby simonw6 days ago|
parent|
[-]
Yeah, there's a reason it still has "preview" and "experimental" in the model names.
reply