OpenAI is rolling out Flex processing, a new API option designed for cost-conscious developers. 😊 This feature cuts prices in half for those willing to accept slower response times and occasional unavailability. It’s perfect for non-urgent tasks like model evaluations or data enrichment.
For the o3 model, Flex processing costs $5 per million input tokens and $20 per million output tokens—half the standard rate. The o4-mini model sees even greater savings, dropping to $0.55 and $2.20 per million tokens, respectively.
This move comes as AI costs rise and competitors like Google release more affordable models. OpenAI also introduces ID verification for developers in certain usage tiers, aiming to curb policy violations.
Flex processing is currently in beta, targeting non-production workloads. It’s a smart trade-off for projects where speed isn’t critical but budget is. 🚀
