Now O1 istheirquote-unquotereasoningmodel, anditscoresreallyhighoncodingandmathbenchmarks.
O1は彼らの推論モデルで、コーディングと数学のベンチマークで非常に高いスコアを出している。
Itfeelssimilartoothermodelsbuthastheabilitytopromptitselfoverandoveragainusing a techniquecalledChainofThoughtwhichgeneratesreasoningtokens, andhastheabilitytosolvemorecomplexproblems, especiallywhenitcomestomathandprogramming.
Yousee, a fewyearsago, ElonMuskoverpaidfor a failingsocialmediawebsitecalledTwitter, sohecouldbanthiscollegekid's accountwhomade a Pythonapptotrackthelocationofhisprivatejet.