AI Briefing
KOSign in

OpenAI models now deployable on Google Cloud; new 12-hour prompt caching and Decisions API cut costs

·2026.10.23 23:43

Key point

New features include a Decisions API for efficient routing, zero-cost voice design, and GPT-Live-1.1 for faster real-time responses.

Details

  • Prompt caching duration extended from 30 minutes to 12 hours, reducing input costs for slow loops despite a 2.5x upfront write cost.
  • Decisions API introduced to provide fast, bounded choices without generating long text blocks, cutting token spend for simple labels or next steps.
  • OpenAI models and Codex are now deployable through Google Cloud Marketplace.
  • GPT-Live-1.1 and Quip rolled out for voice, adding faster responses and better language support.
  • Zero-cost voice design allows building custom voices from text prompts, with charges only for standard audio rates when the model speaks.
  • GPT-6 Pro now supports parallel agents working on separate task pieces simultaneously.
  • Admins can block API projects from using high-cost service tiers via hard project limits.
openaiprompt-cachinggoogle-cloudvoice-aiapi-costsgpt-6-pro
Read originalShare on Threads

This summary was generated automatically by AI. Check the original for the author's claims and context. Copyright belongs to the original author.

Our guide explains how the AI works. Report summary errors, attribution issues, or removal requests via Contact.

Related

OpenAI BlogOpenAI Frontier Models and Codex to Arrive on Google Cloud Model GardenKarrotDanggeun: Cache hit rate drops to 2.3% after GPT-5.6 swap, increasing LLM costs by 3.2xGoogle Cloud AIGoogle Cloud launches OpenAI models in its model gardenOpenAI BlogOpenAI Decisions API generally available with GPT-6 Luna for 10x faster bounded decisionsTLDR AIOpenAI releases official CLI for REST API accessCloudflare BlogCloudflare cuts AI code review costs by 68% to $0.77 via prompt caching and model optimization