OpenAI is launching Flex Processing, a new mode of operation for its API that reduces the cost of using AI models in exchange for possible response delays and periodic service unavailability.
OpenAI is launching Flex Processing, a new mode of operation for its API that reduces the cost of using AI models in exchange for possible response delays and periodic service unavailability.
The new option is available in beta for the recently released OpenAI reasoning models o3 and o4-mini . Flex Processing is aimed at lower-priority and “non-production” tasks, such as model evaluation, data enrichment, and asynchronous workloads — that is, where speed is not critical, writes TechCrunch.
Flex Processing cuts API costs by exactly half. For o3, Flex processing costs $5 per million input tokens (~750,000 words) and $20 per million output tokens, compared to the standard $10 per million input tokens and $40 per million output tokens. For o4-mini, Flex reduces the price to $0.55 per million input tokens and $2.20 per million output tokens from $1.10 per million input tokens and $4.40 per million output tokens.
In an email to customers about the new option, OpenAI also noted that developers in tiers 1-3 of the usage tier hierarchy will need to go through a newly implemented identity verification process to gain access to o3. The tiers are determined by the amount of money spent on OpenAI services. The summaries of o3 reasoning and other models, as well as support for the streaming API, are also closed to verification.
OpenAI previously stated that ID verification is designed to prevent attackers from violating usage policies.
