Openai launches Flex For Artiper and Plate

In an attempt to compete more strongly with artificial intelligence competing companies such as Google, Openai is launched Flexible treatmentAPI option, which provides lower artificial intelligence model prices for the slower response times and “the lack of transverse resources”.

Flex Processing, available in Beta for the recently released O3 and O4-Mini Openai says that thinking models are aimed at low priority and “non -production” tasks such as models assessments, data enrichment, and asymmetric work burdens.

It reduces the API costs exactly half. For O3, Flex $ 5 processing/m input symbols (about 750,000 words) and $ 20/m codes for $ 10/m $ 40/m. For O4-MINI, Flex reduce the price to 0.55 dollar entry codes, $ 2.20/m of icons of input 1.10 dollars/m and $ 4.40/m output symbols.

The launch of Flex Processing comes as The price of Frontier AI continues to go upAnd with competitors launch cheaper and more efficient models. On Thursday, Google has been released Gemini 2.5 flashThinking model that matches or best Deepseek’s R1 In terms of performance at a lower input code cost.

in Email to customers When announced the launch of Flex Flex, Openai also indicated that developers at 1-3 levels of hierarchy for use will have to have a hierarchy to complete The process of verifying the newly provided identity To reach O3. (The levels are determined by the amount of money spent on Openai services.

Openai previously said that verification of identity aims to prevent bad actors from violating their own use policies.