Live data from Hacker News

Degraded performance for multiple models

status.claude.com

21–30 of 149 posts

Re: Degraded performance for multiple models

#25

ive found degraded performance on models larger than 4.7. i assume its model damage from overly self righteous post training resulting in false/feigned balance imported into any long running complex task. wish i was joking.

Aren't the model weights frozen?

Model competence is an interaction of weights, system prompt, and harness.
Post reply on HN