Google (@Google) “Most models are built to generate text, creative content, or reason step-by-step”

2026.11.11 18:26

Most models are built to generate text, creative content, or reason step-by-step. But not every task needs a written answer. When a workflow or agent just needs to make a quick call, the Decisions API returns calibrated confidence scores (150–250 ms end-to-end) at $0.10 per 1M input tokens, making it nearly 5x more cost-efficient across the shared benchmark suite than Gemini 3.5 Flash-Lite with the Interactions API.

显示更多