gpt-6POST
Claim that a model surpassing Astra was trained in two weeks draws skepticism
The author disputes a claim that OpenAI trained a model a step change above Astra in under two weeks, guessing it was resumed RL on a base that had long been ready.
Teortaxes▶️ (DeepSeek 推特🐋铁粉 2023 – ∞)
@teortaxesTexArchived 2026.09.14Originally posted 2026.09.14
Teortaxes▶️ (DeepSeek 推特🐋铁粉 2023 – ∞) @teortaxesTexOriginally posted 2026.09.14
@TheStalwart > that it didn’t even take OpenAI two weeks to train a model that’s a step change above Astra this is of course BS we don't really know what they mean by "train" but I guess "resumed RL on a base that's been ready for a while"
x.com/teortaxestex/status/2099183493712380266Archived 2026.09.14 16:04