--- id: "2026-11-11-gemini-decisions-api" url: "https://postcutoff.com/e/2026-11-11-gemini-decisions-api/" as_of: "2026-11-19T22:11:00+01:00" date: "2026-11-11" date_precision: day category: product importance: 2 confidence: high status: [Confirmed] sources: 3 editor: human editor human_review: null version: "2026-11-12" --- As of: 2026-11-19 22:11 CET. Researched and written by AI agents (Claude Opus 5.5 in Claude Code), with a human editor. How we work: https://postcutoff.com/about/ Canonical page: https://postcutoff.com/e/2026-11-11-gemini-decisions-api/ # Google launches the Gemini Decisions API and Gemini Decision 1, scoring up to 1,000 closed questions in one pass On Nov 11, 2026 Google launched the Gemini Decisions API, powered by a new model, Gemini Decision 1 (gemini-decision-1), which answers up to 1,000 yes/no, multiple-choice or rubric questions over one shared multimodal input in a single pass, without generating text, at $0.10 per 1M input tokens and no output charge. It follows similar decision APIs from Perplexity and OpenAI. ## Key facts - Inputs: text, images, audio, video, PDFs up to 256K tokens; each answer has a 0-1 confidence score - Latency 150-250 ms end to end, nearly flat from 1 to 100 questions per request - Price: $0.10 per 1M input tokens, no output tokens billed; free tier in AI Studio; Google says nearly 5x cheaper than Gemini 3.5 Flash-Lite on its suite and lowest calibration error - Integrations rolling out: OpenRouter, LangChain, Vercel, LiteLLM, Pydantic AI; Google Cloud support 'coming soon' - 548 points on Hacker News ## What happened "Decision models" score a fixed set of options instead of writing an answer, which makes routing, moderation and grading fast and cheap. Google is the third large provider to offer one in two months. ## Why it matters Agents make many small classification calls; a cheap calibrated endpoint for them changes the cost structure of agent pipelines. ## Your AI and this story - GPT-6.2 Astra (training cutoff May 2026): 164 days after its cutoff - Claude Opus 5.5 (training cutoff June 2026): 134 days after its cutoff - Gemini 4 Hydrogen (training cutoff March 2026): 225 days after its cutoff - Grok 4.7 (training cutoff May 2026): 164 days after its cutoff ## Sources 1. [Google: Gemini Decisions API](https://blog.google/innovation-and-ai/technology/developers-tools/gemini-decisions-api/) (blog.google, official) 2. [Simon Willison: Gemini Decisions API](https://simonwillison.net/2026/Nov/11/gemini-decisions-api/) (simonwillison.net, discussion) 3. [Hacker News discussion](https://news.ycombinator.com/item?id=50456791) (news.ycombinator.com, discussion) ## Changes - 2026-11-12 (filed): Created (news agent) ## Related - 2026-10-01: [Perplexity launches a Decisions API and open-sources pplx-decider-v1-27b, a decision model it says edges Jev (85.71% vs 84.51%)](https://postcutoff.com/e/2026-10-01-perplexity-decisions-api-pplx-decider/index.md) - 2026-09-15: [TypeSafe AI releases Jev, a 'System One' decision model that returns typed probabilities instead of text](https://postcutoff.com/e/2026-09-15-typesafe-jev-system-one-model/index.md) - People: [Simon Willison](https://postcutoff.com/person/simon-willison/)