Sam Altman and team kicked off the company’s “12 Days of OpenAI” event Thursday with a live stream to debut the fully functional version of its 01 reasoning model, as well as a new subscription tier called ChatGPT Pro. But to gain unlimited access to these new features and capabilities, you’re going to need to shell out an exorbitant $200 per month.
The 01 model, originally codenamed Project Strawberry, was first released in September as a preview, alongside a lighter-weight o1-mini model, to ChatGPT-Plus subscribers. o1, as a reasoning model, differs from standard LLMs in that it is capable of fact-checking itself before returning its generated response to the user. This helps such models reduce their propensity to hallucinate answers but comes at the cost of a longer inference period and slower response.
Still, these models can “deduce” answers to complex queries in technical fields like science, coding, and math faster than humans can. According to OpenAI’s internal tests, the new version of o1 reduces “major errors” on “difficult real-world questions” by 34% over its preview version.
The new model is available to ChatGPT Plus and Team users starting today through the chatGPT website’s model selection tool. Enterprise and Edu subscribers will get access beginning next week. While you won’t initially be able to upload files or access the web with the new o1, OpenAI says it plans to add support for those and other features in the coming months.
The company also debuted its new all-inclusive subscription tier, the $200/month ChatGPT Pro. With it, users will gain unlimited access to all of OpenAI’s models, including the GPT-4o, the full version of o1, and Advanced Voice Mode. Pro subscribers will also get exclusive access to o1 Pro mode, which “uses more compute for the best answers to the hardest questions,” per the company.
“In evaluations from external expert testers, o1 pro mode produces more reliably accurate and comprehensive responses, especially in areas like data science, programming, and case law analysis,” OpenAI told TechCrunch. “Compared to both o1 and o1-preview, o1 pro mode performs better on challenging machine learning benchmarks across math, science, and coding. In particular, we saw a 75 percent reduction in errors for easier coding competition questions more reflective of everyday programming queries.”
OpenAI has been under considerable pressure from its investors in recent months to reduce the rate at which the company — which at one point was estimated to be spending $700,000 per day to operate ChatGPT — has been losing money. The company recently floated the idea of incorporating advertising into its chat results as a means of generating revenue and had previously suggested charging Enterprise customers upward of $2,000 per month for an elite subscription tier. And, according to The New York Times, OpenAI could increase the rates on Plus-tier subscriptions up to $44 per month by 2029. Whether users are willing to shell out that much cash for a product that is still so consistently, “confidently wrong” as ChatGPT remains to be seen.