News for hackers. Stuff that matters.

A comment on GPT-5.6. Parent comment

Re:GPT-5.6 by coldtea on Friday July 10, 2026 @10:16AM (#48860290)
"Their performance depends solely on the model training before release and how well you curate the context you feed it".

Wrong. The face-saving backtracking doesn't change that.

[ Parent | on HN ]
Re:GPT-5.6 by daveguy on Friday July 10, 2026 @10:18AM (#48860311)
The models do not get better until a new one is released. And we are already at diminishing returns. So sorry. Also sorry you don't know the difference between a model and a context, harness, router, or cache.
[ Parent | on HN ]