Google launches Gemini Decisions API with Gemini Decision 1 model at $0.10 per 1M input tokens
Google released Gemini Decision 1, its first decision model, and launched the Gemini Decisions API through the Gemini API and Google AI Studio. The model scores closed-set choices — yes/no, multiple choice up to 255 options, and rubric scoring — across up to 1,000 questions per request, handling text, images, audio, video and PDFs in a single forward pass. It returns probabilities and confidence scores at 150–250 ms end-to-end latency, priced at $0.10 per 1M input tokens with no output token charges. Third-party testing placed it at 66 on the Artificial Analysis Decision Index v2.1, seven points behind the leaders.
IllustrationEditorial responsibility
- No named human review is recorded for this page.
- Reports are grouped by semantic similarity and deterministic rules. Language models may assist titles, summaries, translation and cross-source analysis; the page reads the event directly, while its address stays stable when the title changes.
- Summary covers the current reports
Cross-source coverage
Reporting timeline
RT @OfficialLoganK: Introducing Gemini Decision 1, our new frontier decision model in the Gemini API. It is $0.10 per 1M input tokens wit…
Logan Kilpatrick, posting on X, announced the introduction of Gemini Decision 1, described as a new frontier decision model available in the Gemini API. The post states the model is priced at $0.10 per 1M input tokens. The announcement presents the model as a new offering within Google's Gemini API lineup, with the pricing detail serving as the primary concrete information disclosed in the post. No further technical specifications, availability dates, or performance benchmarks were included in the visible text.
Read sourceGoogle Launches Gemini Decisions API for Fast Multimodal Probabilistic Choices
Google DeepMind introduced the Gemini Decisions API, powered by the new Gemini Decision 1 model (gemini-decision-1), available through the Gemini API and Google AI Studio. The endpoint handles closed-set decisions such as tool selection, policy checks, model routing, and context grading. Users can submit one multimodal input (text, images, audio, video, PDFs up to 256K tokens) with up to 1,000 independent questions, mixing yes/no, multiple-choice, and rubric-scoring types. Instead of token-by-token generation, the model scores candidate options in a single forward pass, returning probabilities and confidence scores at 150-250 ms end-to-end latency, priced at $0.10 per 1M input tokens with zero output tokens billed. Google claims leading benchmark performance with the lowest calibration error and says it is nearly 5x cheaper than Gemini 3.5 Flash-Lite with the Interactions API. Internal uses include model routing in Gemini Notebook, rapid design iteration in Stitch, emoji selection in CC, and a safety layer for MCP tools in Search.
Google's very first decision model is here, and it's on OpenRouter from day one. Gemini Decision 1 answers yes/no, multiple choice, and score questions with a probability, for $0.10/M input and free output. https://t.co/x95JfrtOEf
Google has released Gemini Decision 1, described as its first decision model, and made it available on OpenRouter from launch day. According to the announcement, the model answers yes/no, multiple choice, and score questions by returning a probability. Pricing is listed at $0.10 per million input tokens with free output. The post provides no further details on model architecture, benchmarks, or availability beyond OpenRouter.
Read sourceShow 2 older updatesHide older updates
Google Releases Gemini Decision 1, Scoring 66 on Artificial Analysis Decision Index
Google released Gemini Decision 1, its first decision model, scoring 66 on the Artificial Analysis Decision Index v2.1, in line with Elephant 9B and Jev 1.14 and 7 points behind leaders Drex Max v1.1 and GLiDE v1.2, which both score 73. It is priced at $0.10 per 1M input tokens and costs $345 per 1M Tasks across the Index, around three times higher than Drex Max v1.1 ($103) due to higher pricing. The model leads knowledge and reasoning benchmarks, scoring 66 on AA-Omniscience-Decision and 49 on HLE-MultiChoice, but trails on the AA-Paperclip classification and extraction benchmark, answering all 15 questions correctly on 33% of documents versus 63% for Drex Max v1.1. It shows strong calibration for reasoning tasks with a Brier score of 0.18 across the Index, between Drex Max v1.1 (0.16) and Jev 1.14 (0.19), and a 0.20 Brier score on HLE-MultiChoice among the lowest in the Index.
Read sourceGoogle Launches Gemini Decisions API Powered by New Gemini Decision 1 Model
Google announced the launch of the Gemini Decisions API, powered by its new Gemini Decision 1 model in Google AI Studio. Rather than generating text token by token, the model is designed for closed-set choice scoring across up to 1,000 questions spanning text, images, audio, video, and PDFs in a single pass. It supports three decision types: yes/no, multiple choice with up to 255 options, and rubric scoring. Pricing is set at $0.10 per 1M input tokens with no charge for output tokens, which Google says is roughly 5x cheaper than Flash-Lite. The company claims the model leads on several benchmark evaluations with the lowest calibration error. Google teams are already building with it, including Stitch Rapid, which is shown designing a full product flow in seconds.