Hacker News Reader

Comment by rockdoe | Hacker News Reader

the surprisingly capable Haiku 5.5

Destroyed by Anthropics crazy decision to price it x5 after 100k context.

do people really use luna post 100k? small models generally seems to be hallucinating hard for me once you are over that threshold.

On purpose? I was using qwen3.8-27b the other day and very quickly ended up burning through half of the context (it has 128k total) pretty quickly. Not because of my directions, but simply because the codebase was very large.

Is there a straightforward way to check the current context window usage in the web chat interfaces, btw?

I don't know, I usually use Pi.

In Claude it used to be the circle on the lower bottom, but now it's a bit more hidden and you have to expand the sidebar.

256k seems sweet spot. Even large codebase can be handled you just has to map out areas for agent. Anything past 256k usually exception than norm to be.

At least this my discovery when I did evals on my reviewers skills. 10k LOC is high end for high effort or around 150k context using opus. You can push towards 15k but that’s on max effort with about 400-500k ctx.

Fast models like haiku do not even need 1M ctx in my view

Yeah codebase structure is very important. Admittedly, by large I mean average size but very poorly structured with a spaghetti (and sometimes cyclical) import graph.

Due the the size of the default system prompt you get over 100k really in easy in Claude, and because that is fixed 200k is really almost 3x more room to work with, not 2x.

200k also used to be the limit for Sonnet for a long time.

If Haiku 5.5 also had 200k, it'd actually be usable for us. As it is, it's Luna or Sol.

Packaging & pricing seem to be two things Anthropic really struggles with

Reply on news.ycombinator.com