4 ms·
it feels like openai are at a ceiling with their models, codex1 seems to be another RLHF derivative from the same base model. You can see this in their own sel
by strangescript 1y ago
it feels like openai are at a ceiling with their models, codex1 seems to be another RLHF derivative from the same base model. You can see this in their own self reported o3-high comparison where at 8 tries they converge at the same accuracy.
It also seems very telling they have not mentioned o4-high benchmarks at all. o4-mini exists, so logically there is an o4 full model right?
- aorobin 1y agoSeems likely that they are waiting to release o4 full results until the gpt-5 release later this year, presumably because gpt-5 is bundled with a roughly o4 level reasoning capability, and they want gpt-5 to feel like a significant release.
- losvedir 1y agoDo you still think there will be a gpt-5? I thought the consensus was GPT-5 never really panned out and was released with little fanfare as 4.1.
- aorobin 1y agoYeah, just last month Altman said gpt-5 is coming in a few months, and betting/prediction sites are expecting it this year, probably in the summer.
- brookst 1y agoMarketing names aren’t really connected to product generations. We might target v3 of a product for a date and then decide it’s really 2.4, doesn’t mean we won’t market something as v3 later.