gpt-6IMAGE
Astra saturates new physics benchmark CritPt right out of the gate
The author notes that Astra saturated a new physics benchmark immediately, crediting OpenAI's strong RL pipeline and expecting Fable 5.5 to saturate it too.
Teortaxes▶️ (DeepSeek 推特🐋铁粉 2023 – ∞)
@teortaxesTexArchived 16 days agoOriginally posted 2026.10.15


Teortaxes▶️ (DeepSeek 推特🐋铁粉 2023 – ∞) @teortaxesTexOriginally posted 2026.10.15
Physics problems. I guess this would have been a nice new CritPt, if Astra didn't saturate it right out of the gate. OpenAI does have a ferociously strong RL pipeline. Fable 5.5 will probably saturate it too.
This post quotes another post: x.com/pratikrath15/status/2110225213401206892
x.com/teortaxestex/status/2110440866200641847Archived 2026.10.15 10:50