OpenAI The company's first "reasoning" models, o1 and o1-mini, are now available to Enterprise and Edu users, who can enter up to 50 messages per week.
According to the introduction, OpenAI o1 series models have higher understanding ability, unlike traditional models, o1 series models will think deeply before answering the question, generating the inner chain of thought before giving the answer, so the series models can deal with more complex tasks than the previous models, especially good at solving difficult problems in the fields of science, programming and mathematics.
OpenAI notes that while the GPT-4o is still the best choice for most tasks today, the o1 series performs even better at handling complex, problem-solving-oriented tasks such as research, strategy, programming, math, and science.
However, because the o1 model "consumes more computational resources," OpenAI claims that the model series will initially be available only to a limited number of Plus/Team/Enterprise/Edu users.The o1 model can be used up to 50 times per week, while the o1-mini model can be used up to 50 times per day..
On the API side, the price of access to the relevant models is high, with o1's input tokens costing $1,000,000 per million. 15 USD (IT Home Note: Currently about 107 RMB), and the output word dollar cost is per million 60 The o1 model can be called up to 500 times per minute and the o1-mini model can be called up to 1000 times per minute when called through the API.