OpenAI launches Flex processing for cheaper, slower AI tasks

To compete more aggressively with rival AI companies like Google, launching OpenAI Flex processingAn API option that provides the cost of using low AI models in exchange for slow response time and “Occasionally resource unavailable”. “

Flex Processing, which is available in the recently published Bita of Openai O3 and O4-Mini OpenAI says that rational models look at lower-priority and “non-production” functions such as model evaluation, data enrichment and asynchronous work stress.

It reduces the API expenses exactly half. For O 3, flake processing is $ 5/m input token (~ 750,000 words) and 20/m output token vs. standard $ 10/m input token and $ 40/m output token. For O4-Mini, the flake price drops $ 1.10/m input token from $ 2.20/m output token and $ 4.40/m output token.

Comes as the introduction of flax processing The Frontier AI’s Price continues to climbAnd rivals publish such cheap, more skilled budget-based models. Thursday, Google Roll Out Gemini 2.5 flashAn argument model which match or best DIPSECR R1 In the case of performance at low input token costs.

A Email to customers OpenAI also announced that Flex Papering was launched The process of verification of newly introduced ID To access the O3. (Levels are determined by the amount of money spent on OpenAI services)

Open has earlier said that the purpose of the ID verification is to prevent the use of bad actors from violating the use of bad actors.