OpenAI has launched ChatGPT Pro, a $200 per month subscription plan that provides unlimited access to the full version of o1, its new large reasoning model, and all other OpenAI models. The toolkit includes o1-mini, GPT-4o and Advanced Voice. It also includes the new o1 pro mode, “a version of o1 that uses more compute to think harder and provide even better answers to the hardest problems,” OpenAI explains, describing the high-end subscription plan as a path to “research-grade intelligence” for a way for scientists, engineers, enterprise, academics and others who use AI to accelerate productivity.

“We think the audience for ChatGPT Pro will be the power users of ChatGPT — those who are already pushing the models to the limits of their capabilities on tasks like math, programming, and writing,” Jason Wei, a member of OpenAI’s technical staff, said during a Thursday press conference reported by TechCrunch.

OpenAI began previewing o1 in September as part of a new series of LRMs that can correct their own mistakes, taking a bit longer to process queries but providing more reliable responses. The o1 pro mode produces the most reliably accurate and comprehensive responses yet, “especially in areas like data science, programming, and case law analysis,” OpenAI notes in an announcement.

Across machine learning benchmarks, o1 pro mode performs better than both o1 and o1-preview on math, science and coding, OpenAI says.

“Additionally, o1 can reason about image uploads now (this wasn’t possible during the preview),” reports TechCrunch. VentureBeat calls the ability to analyze images “a hugely helpful feature” that allows users to upload images so the AI chatbot can respond to them, “giving them detailed plans on how to build a birdhouse entirely from a single candid photo of one, for one fun example.”

VentureBeat shares another, “potentially more serious and impressive example” shared on X of o1 “helping design data centers from sketches.”

The o1 release model also “has been trained to be ‘more concise in its thinking’ to improve response times,” TechCrunch reports, citing OpenAI internal testing data indicating “o1 reduces ‘major errors’ on ‘difficult real-world questions’ by 34 percent compared to the preview version.”

With ChatGPT Pro’s launch, o1 officially transitions out of preview, becoming a foundational part of the ChatGPT framework.

And you don’t have to be a ChatGPT Pro subscriber to use it, TechCrunch points out, noting “all paid ChatGPT users can access o1.” Athough lower tier subscribers get “limited access,” as explained in OpenAI’s GPT pricing chart.