Archive — US — techcrunch
OpenAI launches Flex processing for cheaper, slower AI tasks
OpenAI has introduced Flex processing, a cheaper API option that reduces costs for lower-priority AI tasks in exchange for slower response times and occasional unavailability. This move is aimed at competing with rival AI firms like Google. Flex processing decreases costs by half and is in beta for OpenAI's o3 and o4-minireasoning models, launched amid rising frontier AI prices and increasing competition from models like Google's Gemini 2.5 Flash.
Key points
- OpenAI launches Flex processing to lower API costs for non-production AI tasks.
- Flex processing reduces costs by 50% for o3 (from $10/M to $5/M input tokens) and o4-mini (from $1.10/M to $0.55/M).
- Flex targets lower-priority tasks like model evaluations and data enrichment.
- The launch responds to rising AI model prices and competitors like Google releasing cheaper models.
- Developers in usage tiers 1-3 must complete ID verification to access the new pricing and features.
Why it matters
- As of 2025, budget-friendly AI models are becoming crucial in a competitive landscape dominated by companies like Google and emerging firms.
- Flex processing aligns with the trend of making AI more accessible and cost-effective for various applications.
OpenAI's Flex processing is a strategic response to competitive pricing in the AI market, making AI more accessible while addressing the need for budget-oriented solutions in an increasingly costly environment.