OpenAI Launches Flex Processing to Reduce AI Costs for Less Urgent Tasks
OpenAI has introduced a new feature called Flex processing, aimed at helping users save on costs associated with AI usage. Although this option provides lower prices, it comes with some trade-offs, including slower performance and potential access issues. This initiative is part of OpenAI’s strategic response to the rising competition, particularly from companies like Google, that are developing more affordable and faster AI models.
Flex processing is designed to accommodate tasks that are less time-sensitive, such as testing models, enriching datasets, or carrying out background processes. Currently, this feature is in beta and is available for two of OpenAI’s most recent reasoning models: o3 and o4-mini.
Cost Savings for Routine Tasks
One of the hallmarks of Flex processing is its ability to cut API usage costs by 50%. Users will pay only USD $5 for every million input tokens (approximately 750,000 words) and USD $20 for million output tokens when utilizing the o3 model. In contrast, the standard pricing for these tokens is USD $10 and USD $40, respectively. When employing the o4-mini model, the rates further drop to USD $0.55 for million input tokens and USD $2.20 for million output tokens, compared to regular prices of USD $1.10 and USD $4.40.
While these significant savings are appealing, users should be aware of the accompanying drawbacks. Flex processing generally results in slower response times and may cause delays if resources are not immediately available. However, for tasks that do not necessitate immediate results, this option can help maintain budgetary constraints while still offering access to advanced AI capabilities.
Timing and Need for Flex Processing
The introduction of Flex processing is timely, as rising costs associated with AI development drive companies to look for budget-friendly solutions. Recently, Google launched Gemini 2.5 Flash, which is a cost-effective AI model that is competitive with offerings from firms like DeepSeek. It delivers great performance at a lower cost per input token.
With Flex processing, OpenAI aims to make it easier for users to conduct experiments or process significant amounts of data without incurring high expenses. This feature is particularly beneficial for users who want access to powerful AI tools without the need for real-time computations.
ID Verification for Select Users
In conjunction with the rollout of Flex processing, OpenAI has initiated an ID verification requirement for some developers. This regulation applies to users in tiers 1 through 3, which are determined based on the amount spent on OpenAI services. If you are in one of these categories and wish to access the o3 model or particular features—like reasoning summaries or the streaming API—you will need to verify your identity.
OpenAI has stated that this measure is intended to prevent the misuse of its tools and ensure compliance with company policies. As artificial intelligence continues to grow in capability, organizations are taking additional precautionary steps to manage who can utilize these powerful systems and how they are used.
For individuals managing a project, experimenting with AI, or running a business, Flex processing presents an economical alternative, especially for tasks that don’t require immediate responses. With its competitive pricing and enhanced flexibility, it could align perfectly with your upcoming AI-enhanced projects.
