OpenAI has launched a extra highly effective model of its o1 “reasoning” AI mannequin, o1-pro, in its developer API.
According to OpenAI, o1-pro makes use of extra computing than o1 to supply “consistently better responses.” Currently, it’s solely accessible to pick builders — those that’ve spent a minimum of $5 on OpenAI API providers — and it’s dear. Very dear.
OpenAI is charging $150 per million tokens (~750,000 phrases) fed into the mannequin and $600 per million tokens generated by the mannequin. That’s twice the value of OpenAI’s GPT-4.5 for enter and 10x the value of standard o1.
OpenAI is betting that o1-pro’s improved efficiency will persuade builders to pay these princely sums.
“O1-pro in the API is a version of o1 that uses more computing to think harder and provide even better answers to the hardest problems,” an OpenAI spokesperson informed TechCrunch. “After getting many requests from our developer community, we’re excited to bring it to the API to offer even more reliable responses.”
Yet early impressions of o1-pro, which has been accessible in OpenAI’s AI-powered chatbot platform ChatGPT for ChatGPT Pro subscribers since December, weren’t extremely constructive. The mannequin struggled with Sudoku puzzles, customers discovered, and was tripped up by easy optical phantasm jokes.
Furthermore, sure OpenAI inside benchmarks from late final yr confirmed that o1-pro carried out solely barely higher than the usual o1 on coding and math issues. It did reply these issues extra reliably, nevertheless, the benchmarks discovered.
