OpenAI launches Flex processing for cheaper, slower AI tasks

Google, as Openai offers to compete more aggressively with its rival AI companies Flexible processingAn API option that provides the prices using a low AI model in exchange for more slow response periods and “lack of random resource”.

Flex processing available in Beta because Openai releases recently O3 and O4-Mini Model assessments, information enrichment and asynchronous workloads, information richness and asynchronous workloads are directed to “non-produced” positions.

API reduces costs in full half. For O3, FLEX processing is 5 / m / 750,000 words) and 20 / m / m output signs and $ 10 / m Input Tokens and 40 / M / M Product Tokens. For O4-mini, the Flex price brings $ 0.55 / m in access tokens and $ 1.10 / m from $ 1.10 / m to $ 1.10 / m and $ 4.40 / m from $ 4.40 / m.

Comes as the launch of flex processing The border price continues to climb the AIAnd as opponents release cheaper, more efficient budget-oriented models. On Thursday, Google spread Twins 2.5 Flasha model of thinking or best of the best Deepseek’s R1 In terms of performance in a lower access sign.

One Email customers Announcing the start of the Flex Price, Openai also said that the number 1-3 classes of the stages used in the stages used The newly presented identity verification process To access O3. (Ties are determined by the amount of money spent on Openai services.) Summary of O3 and other models – and other models and flow API support are also behind the check.

Openai previously said that the identity card is designed to avoid violating the rules of use of bad actors.