Artificial intelligence / news brief
OpenAI makes its low-latency Decisions API generally available
- Event
- Published
OpenAI made its GPT-6 Luna-based Decisions API generally available for bounded choices from text, images and structured inputs. It costs $0.10 per million input tokens and adds voice integration, caching and regional processing; OpenAI's up-to-10x speed claim and customer latency examples are not independent benchmarks.
Why it matters
A cheap, constrained decision endpoint gives developers an alternative to full language-model calls for routing, classification and real-time agent control.
Sources
- Build more responsive apps with the Decisions API (external site)OpenAI · primary source ·
- OpenAI makes its Jev competitor widely available as decision model competition heats up (external site)TechCrunch · coverage ·