Re:Claude Opus 5.5
by bpodgursky on Tuesday September 22, 2026 @12:52PM
(#49804335)
Everyone knows both labs have internal models which outperform the frontier. All releases are to match market parity and demand for spend, the rest of the compute is used for training. It's not worth arguing about this.
[ Parent
| on HN ]
Are those powerful models in the room with us right now?
then technically the internal models are the frontier
Re:Claude Opus 5.5
by nextaccountic on Tuesday September 22, 2026 @01:06PM
(#49804552)
Maybe their internal edge dried up in the last months