OpenAI has officially launched Flex processing, a new API option designed to significantly reduce costs for users handling lower-priority AI tasks. We are running this new offering on a beta basis for now. It aims to undercut the o3 and o4-mini reasoning models, vowing to halve API costs for developers seeking to refine their workloads.
Smart flex processing is all about making the experience best-suited to those completing non-production work. This involves things such as performing model assurances, enriching data, and fulfilling asynchronous workloads. Required completion of these kinds of tasks isn’t time sensitive, so these types of tasks are excellent candidates for this innovative and low-cost solution. Flex processing is smartly introduced by OpenAI, giving developers an opportunity to get the most out of their resources, without compromising quality of performance.
The cost of Flex processing in the o3 model is $5 per million input tokens and $20 per million output tokens. This is a huge discount from the regular API rates, which are $10 per million input tokens and $40 per million output tokens. This change leads to a 50% decrease in the cost that users pay when they choose to process using the Flex processing option. For the o4-mini model with Flex processing, developers will be charged $0.55 per million input tokens and $2.20 per million output tokens. This is a huge discount from the usual rates of $1.10 and $4.40, respectively.
This new Flex processing is a natural extension of OpenAI’s goal for increasing access to advanced AI technologies in more affordable and accessible ways. OpenAI provides a cheaper alternative for non-time-sensitive tasks. This model is called Distributed AI and it remarkably enables developers to unlock the power of AI capabilities without incurring prohibitive operational costs.
ID verification became mandatory for developers in tiers 1-3. This is especially important for those who are interested in using the o3 model. This latest new requirement is a part of OpenAI’s new initiative to ensure that its technology is used safely and responsibly.