Google Launches Gemini Decisions API with Gemini Decision 1 Model
On November 12, Google officially launched the Gemini Decisions API, powered by a new frontier decision model, Gemini Decision 1. Both Sundar Pichai and Google AI Studio made the announcement; the product is live in Google AI Studio and available via the Gemini API. Rather than token-by-token text generation, it is purpose-built for closed-ended judgment scenarios.
Confirmed
- Aggressive pricing: $0.10 per million input tokens, output tokens free of charge, no extra fees
- Three decision types: yes/no, multiple choice (up to 255 options), and rubric-based scoring
- Multimodal input: native support for text/JSON, PDF, images, audio, and video, with a 256K-token context window
- A single request can deliver fast, well-calibrated decisions on up to 1000 questions, with end-to-end latency of 150-250ms
- According to Logan Kilpatrick, head of Google AI Studio, Gemini Decision originated as a grass-roots project inside Google DeepMind; the team built a prototype, opened it to internal testing, and went from prototype to launch in roughly 6 weeks
- Per @philschmid (who says he was involved in designing the API), it is built for yes/no, multiple-choice, and rubric-type judgments, returning answers in about 150ms
Why it matters
- The product shifts LLMs from open-ended text generation toward high-throughput, low-cost, structured batch judgment, with pricing and latency metrics directly attractive to content moderation, evaluation, and data labeling applications
- The path from internal DeepMind tool to launch in 6 weeks also showcases a new model of small Google teams rapidly productizing internal projects
2026-11-12 ~ 2026-11-12 · 22 related posts
- Episode 1: Google Launches Gemini Decisions API with Gemini Decision 1 Model(2026-11-12, 22 posts)
- Episode 2: Gemini Decision 1 debuts: tops decision benchmarks but costs 3x rivals(2026-11-12, 6 posts)
Primary sources
- Google launches Gemini Decision 1 decision model at $0.10 per 1M input tokens — OfficialLoganK ·
- Gemini Decision went from DeepMind prototype to shipped API in six weeks — OfficialLoganK ·
- Google ships Gemini Decisions API: 1,000 questions per request at 150-250ms, $0.10 per 1M tokens — _philschmid ·
- Google's Gemini Decisions API returns yes/no, MCQ and rubric calls in 150-250ms — _philschmid · 2026-11-12
- [source] Google launches Gemini Decision 1 decision model at $0.10 per 1M input tokens — OfficialLoganK · 2026-11-12
- Google launches Gemini Decision 1: $0.10/1M input tokens, native multimodal, 256K context — OfficialLoganK · 2026-11-12
- Google officially launches Gemini Decisions API powered by Decision 1 — gaganghotra_ · 2026-11-12
- [source] Gemini Decision went from DeepMind prototype to shipped API in six weeks — OfficialLoganK · 2026-11-12
- Gemini Decisions API pricing: $0.10 per 1M input tokens, free output tokens — patloeber · 2026-11-12
- Google's joke-worthy "Gemini Decision 1" frontier decision model draws meme reactions — danshipper · 2026-11-12
- Google launches Decision Model category with Gemini Decision 1 on Vercel AI Gateway — steren · 2026-11-12
- Google Launches Gemini Decisions API; Swarms Ships Day-1 Integration for gemini-decision-1 — KyeGomezB · 2026-11-12
- Gemini Decisions API returns calibrated, structured choices instead of free text — ColbyHawker · 2026-11-12
12 near-duplicate retellings: _philschmid · Swarooprm7 · _philschmid · osanseviero · osanseviero · patloeber · gaganghotra_ · dair_ai · DynamicWebPaige · dean_rie · rseroter · shensi