OpenAI releases o3-pro, a souped-up version of its o3 AI reasoning model

Open is there Opened O 3-Pro, an AI model that the company claims that it is still the most capable.

O A version of 3-Pro OpenAI O3An argument model that started early this year. Unlike the conventional AI model, rational models work through step by step problems, enables them to perform more reliably in domains such as physics, math and coding.

O 3-Pro ChatzPT Pro and Team users starting from Tuesday, replacing it, replacing it O 1-pro Model. Openi says Enterprise and EDU users will get access after the week. And 3-Pro this afternoon is live in the API developer of the Open.

O 3-Pro costs token per million input tokens and a million output tokens in API. The input tokens are fed to the tokens in the model, while the output token is token that produces the model based on input token.

One million input token is equivalent to about 750,000 words, a little longer than “war and peace”.

“In the evaluation of the expert, reviewers consistently prefer O3-Pro than O3 in original domains, such as science, education, programming, business and writing, especially in each tested department,” “Reviewers have rated a consistently higher and 3-pro for precision, extension, instruction and accuracy.”

According to Openai and 3-PRO equipment, it helps to search the web, analyze the files, logic about visual inputs, using Python, helps to personalize its reactions and many more things. As an error, the model’s reactions usually take longer than O1-Pro, according to Openai.

There are other limitations of O3-Pro. Temporary chats with the model in the ChatzPT have been disabled when the open is solved a “technical problem”. O can not create 3-Pro images. And the canvas, Openai’s AI-powered workspace features, and not supported by 3-Pro.

According to the internal examination of the Openai, on the Plus Side, and 3-POP Popular AI Benchmarks have earned impressive scores. In AIM 2024, which evaluates a model’s math skills, Google’s top-perforcing AI model and 3-Pro score are better, better Gemini 2.5 ProThe O 3-Pro also beats anthropic recently published Stop Fourth Work In GPQA Diamond, a PhD-level science test.