Guardrails removed spam, off-topic, unclear, or duplicate replies.
Ask a question below.
Published answers will appear here.
Interesting question! I tried it on GPT-5.6 Sol (High), available on the Plus plan. I got that the length of the path is <= c_n^n with n being dimension and c_n approaching 2* sqrt(2)=2.828, asymptotically. The uniform over all dimension n result is 3.50427, for n = 4. https://twitter.com/SebastienBubeck/status/2075596982622835006
As Sebastien mentioned, GPT-5.6 Pro gets to c_n of 2.31. The difference 2.828 - 2.31 = 0.518 is the price of using cheaper models (Anyone rich patron wants to fund my lab with Pro plans and a lot of credit?!😃😜).
Interesting question! I tried it on GPT-5.6 Sol (High), available on the Plus plan. I got that the length of the path is <= c_n^n with n being dimension and c_n approaching 2* sqrt(2)=2.828, asymptotically. The uniform over all dimension n result is 3.50427, for n = 4. https://twitter.com/SebastienBubeck/status/2075596982622835006
As Sebastien mentioned, GPT-5.6 Pro gets to c_n of 2.31. The difference 2.828 - 2.31 = 0.518 is the price of using cheaper models (Anyone rich patron wants to fund my lab with Pro plans and a lot of credit?!😃😜).
Guardrails removed spam, off-topic, unclear, or duplicate replies.
Ask a question below.
Published answers will appear here.