> ## Content Index > Fetch the complete content index at: https://www.implicator.ai/llms.txt > Use this file to discover other available public pages before exploring further. # Google Chases OpenAI Into the Decision-Model Market - URL: https://www.implicator.ai/google-launches-gemini-decisions-api-powered-new-gemini/ - Published: 2026-11-12T01:05:29.000Z - Updated: 2026-11-12T01:05:29.000Z - Description: Gemini Decision 1 reads video, audio and PDFs and bills the shared input once per request. A rival charges less for plain text. - Author: Marcus Schuler - Tags: AI News, #auto Google released the [Gemini Decisions API](https://blog.google/innovation-and-ai/technology/developers-tools/gemini-decisions-api/?ref=implicator.ai) on Wednesday, an endpoint that answers closed-set questions instead of writing text. Its new model, Gemini Decision 1, can take up to 1,000 questions about one input in a single pass, at $0.10 per million input tokens with zero output tokens billed. That rate [matches OpenAI's price](https://www.thedeepview.com/articles/why-google-is-chasing-jev-into-decision-models?ref=implicator.ai) for its own decision model, and Jev from TypeSafe remains the cheapest of the three. What Changed - Google released the Gemini Decisions API on Wednesday, powered by its new Gemini Decision 1 model, which scores up to 1,000 questions about one input in a single pass. - It costs $0.10 per million input tokens with zero output tokens billed, the same rate as OpenAI's decision model; TypeSafe prices Jev 1.14 at $0.042 per million. - Google bills the shared input once per request, so 20 checks on a 4,000-token record cost about $0.50 per 1,000 records instead of $8.10 as separate calls. - The independent Artificial Analysis Decision Index v2.1 puts Gemini ninth of 18 at launch. AI-generated summary, reviewed by an editor. [More on our AI guidelines](https://www.implicator.ai/about/). ## Scores, not sentences Rather than generating answer text token by token, Gemini Decision 1 scores candidate options and returns probabilities, with a confidence score from 0 to 1 on every answer. Developers can [mix three question types](https://the-decoder.com/google-joins-the-decision-model-wave-with-gemini-decision-1-scoring-1000-questions-in-a-single-pass/?ref=implicator.ai) in one request: yes/no checks, multiple choice across 2 to 255 options, and grades on an ordered rubric. The shared input can combine text, images, audio, video and PDFs, up to 256K tokens. Google pitches the endpoint for bounded choices like tool selection, policy checks, model routing and context grading. Gemini Notebook already uses it as a router that sends queries to a model tier and thinking level, and Google reports up to 40% lower latency for everyday queries there. Google puts end-to-end latency at 150 to 250 milliseconds for inputs up to 16K tokens, nearly flat from 1 to 100 questions per request. Teams planning to use the full context window or the 1,000-question limit should measure how long their own files and checklists take. ## What one decision costs Google bills the shared input once per request, and each question adds only its own instruction and criteria tokens. Take a [4,000-token record and 20 checks](https://venturebeat.com/technology/googles-gemini-decision-1-model-scores-up-to-1-000-questions-in-one-api-call?ref=implicator.ai) that add 50 tokens apiece. Sent as one 5,000-token request, that costs about $0.50 per 1,000 records at Google's rate. Twenty separate calls that each resend the record would cost $8.10 per 1,000 records, before any retries or follow-up calls. FREE WEEKDAY MORNING BRIEFING Track the price fight over AI decision models. The Implicator Morning Briefing filters the AI news cycle to the stories worth your attention and explains their consequences. From San Francisco, every weekday at 4:45 a.m. Pacific, 7:45 a.m. Eastern. Email address Send me tomorrow’s briefing Check your inbox for the confirmation link. About five minutes. No hype. No spam. Google reports that its shared tests cost $0.88 with Decision 1 and $4.14 with Gemini 3.5 Flash-Lite through the Interactions API. The same comparison put Jev at $0.68. ## Cheaper options for plain text TypeSafe prices Jev 1.14 at $0.042 per million input tokens with free output. Jev takes text and structured JSON, and its opt-in image support lets developers attach a PNG or JPEG. OpenAI made its Decisions API generally available on Oct. 22, with GPT-6 Luna as the sole supported model. It charges $0.10 per million uncached input tokens and $0.01 per million for cache reads. That gives teams an option when they repeatedly send the same policy manual or product catalog across separate requests. The case for Gemini starts with what goes into the request. A team handling recordings, footage and PDFs can evaluate those files together, while a team routing short text tickets has cheaper hosted options. ## An outside scorecard In Google's launch benchmarks, Gemini Decision 1 led five of the seven tests that all four models ran. It also had the lowest calibration error in Google's comparison, 0.0250, roughly half Jev's 0.0494. Know someone who'd find this useful? [✉️ Email it to a friend in one click](mailto:?subject=A%20newsletter%20I%20think%20you%27d%20like&body=This%20is%20one%20of%20maybe%20three%20newsletters%20I%20actually%20read.%20The%20rest%20just%20pile%20up%2C%20unread%2C%20judging%20me.%0A%0AAnd%20yes%2C%20this%20email%20mostly%20wrote%20itself%2C%20which%20is%20a%20little%20on%20the%20nose%20for%20an%20AI%20newsletter.%20Doesn%27t%20make%20it%20wrong.%20implicator.ai%20is%20good.%0A%0ASubscribe%20free%3A%20https%3A%2F%2Fwww.implicator.ai%2Fsubscribe%2F%3Futm%5Fsource%3Dnewsletter%26utm%5Fmedium%3Dforward%26utm%5Fcampaign%3Demail%5Fforward), or they can [subscribe free here](https://www.implicator.ai/subscribe/?utm%5Fsource=newsletter&utm%5Fmedium=forward&utm%5Fcampaign=forward%5Fto%5Fcolleague). The independent Artificial Analysis Decision Index v2.1 puts Gemini ninth of 18 at launch. On AA-Paperclip, which asks 15 questions about each résumé, research paper, invoice or receipt, Gemini got every answer right on 33% of documents. Jev reached 35.5% and Nace AI's Drex Max v1.1 reached 63%. A workflow that requires every field to be correct will care about that all-or-nothing measure. Simon Willison [tested the API on his weblog](https://simonwillison.net/2026/Nov/11/gemini-decisions-api/?ref=implicator.ai), passing his video of a stomping robot to it with four yes/no questions. "Jev and OpenAI both return 1.0 or 0.0 for yes/no questions, but Gemini insists on being a little less confident than that," he wrote. Frequently Asked Questions What is the Gemini Decisions API? It is a Google endpoint for closed-set decisions, powered by the Gemini Decision 1 model. Instead of generating answer text token by token, the model scores candidate options and returns probabilities and confidence scores. It supports yes/no checks, multiple choice across 2 to 255 options, and grades on an ordered rubric. How much does Gemini Decision 1 cost? Google charges $0.10 per million input tokens and $0 for output tokens. The shared input is billed once per request, and each question adds only its own instruction and criteria tokens. A no-cost tier is available in Google AI Studio for prototyping. Is Gemini Decision 1 cheaper than Jev? In Google's own shared tests, the run cost $0.88 with Decision 1 and $0.68 with Jev. What inputs can it handle? The model can combine text, images, audio, video and PDFs in one shared input of up to 256K tokens. Jev takes text and structured JSON with opt-in image support. How fast is it? Google puts end-to-end latency at 150 to 250 milliseconds for inputs up to 16K tokens, nearly flat from 1 to 100 questions per request. Teams using the full context window or the 1,000-question limit should measure their own files and checklists. AI-generated summary, reviewed by an editor. [More on our AI guidelines](https://www.implicator.ai/about/). [OpenAI offers to pay future power bills; hacking AI rarely strays in testsIMPLICATOR .ai Morning Briefing · From San Francisco Friday, November 6, 2026 9 stops From San Francisco 1 The Editorial Morning, humans. AI is getting a lessonThe Implicator![](https://www.implicator.ai/content/images/2026/11/20261105-191412-nl_council_power.webp)](https://www.implicator.ai/openai-power-bills-gpt-6-cyber-sony-ps5/) [LLM Meter — Week of Nov 5, 2026\---CLAUDE--- score: 90 trend: up change: 1 + Anthropic's amended IPO filing reports third-quarter revenue of about $17 billion and adjusted operating profit of $2.5 billion. + Opus 5.5 leads SemiAnalyThe Implicator](https://www.implicator.ai/llm-meter-week-of-nov-5-2026/) [OpenAI's Hacking Model Learns to Stay in Its LaneOpenAI released GPT-6 Cyber, a model trained to stay within the scope researchers authorize, on Thursday, Nov. 5, to a select group of customers in Daybreak Red, its access tier for vetted offensive sThe Implicator![](https://www.implicator.ai/content/images/2026/11/20261105-122659-openai_cyber_lane.webp)](https://www.implicator.ai/openai-gpt-6-cyber-stays-in-scope/)