Gemini Decisions API 快速校准多模态决策

查看产品与平台中的全部内容
查看创新与 AI 中的全部内容
查看产品与平台中的全部内容
Gemini Decisions API:快速、经过校准的多模态决策
Gemini Decisions API:快速、经过校准的多模态决策
推出 Gemini Decisions API,由我们全新的 Gemini Decision 1 模型驱动。可在文本、图像、音频、视频和 PDF 上,以快速单次处理方式评估最多 1,000 个是/否题、选择题和评分标准评分题。
Google DeepMind 产品经理
Google DeepMind 首席科学家
您的浏览器不支持 audio 元素。
我们推出 Gemini Decisions API,帮助您在文本、图像、音频、视频和 PDF 中快速做出概率性选择。该工具可在单次处理中计算多个问题的置信度分数,为路由查询、内容评分或检查策略规则提供了一种高性价比的方式。您今天就可以开始构建:在 Google AI Studio 中获取 API 密钥,并查看我们面向 Python、JavaScript 和 REST 的文档指南。
摘要由 Google AI 生成。生成式 AI 仍属实验性。
今天,我们推出 Gemini Decisions API,这是一个新端点,用于由我们全新的 Gemini Decision 1 模型(gemini-decision-1)驱动的快速闭集决策,可通过 Google AI Studio 中的 Gemini API 使用。
Decisions API 为您提供专门构建的原语,用于工具选择、策略检查、模型路由和上下文评分等有界选择。您可以传入一个跨文本、图像、音频、视频和 PDF 的多模态输入(最多 256K Token),以及最多 1,000 个独立问题。Gemini Decision 1 不会逐 Token 生成答案文本,而是在单次前向传播中对您的候选选项进行评分,返回概率和置信度分数,端到端延迟为 150-250 ms,每 1M 输入 Token 收费 $0.10,输出 Token 不计费。
强劲性能与校准
在涵盖智能体路由、推理、排序和多模态理解的基准测试中,Gemini Decision 1 在整个测试套件中领先,并具有最低的校准误差。
Gemini Decision 1 无需逐 Token 解码即可为各选项打分,模型计算的中位延迟保持在 50-100 ms(端到端 150-250 ms),可处理多模态输入以及最长 16K Token 的上下文;当单个请求中的问题数从 1 增至 100 时,延迟几乎保持不变。由于不计费输出 Token,在共享基准测试套件上,搭配 Decisions API 的 Gemini Decision 1 比搭配 Interactions API 的 Gemini 3.5 Flash-Lite 便宜近 5 倍。
单次处理即可跨三种决策类型完成原生多模态评估
Gemini Decision 1 基于 Gemini 原生的多模态架构构建,可在单次前向传播中直接处理文本、图像、音频、视频和 PDF。你可以在同一份共享输入中组合多种模态,例如将通话录音与账户元数据配对,或将扫描版 PDF 理赔单与损坏照片配对,并可调节图像和视频输入的分辨率,以控制视觉细节和 Token 成本。
你可以在单个请求中针对这份共享输入评估多个问题,混用是/否判断、多选题和评分量表。每个答案都附带置信度分数(0 到 1),这样你的代码就能把模糊情形交给兜底方案或人工审核:
noul(是/否):估计某个是/否条件成立的概率,可选填判定为真与判定为假的描述标准。
choice(多选题):从 2 到 255 个选项中选出答案,返回胜出项以及每个候选选项的归一化概率。
score(有序评分量表):按有序评分量表对输入打分(从 0 开始),返回连续的加权分数以及各档位的概率。
input 字段使用与 Interactions API 相同的 Content 块格式。下面演示如何在单个请求中,用全部三种问题类型来核查一张茶水间台面的照片以及一段文字说明:
from google import genai client = genai.Client(http_options={"api_version": "v1beta"}) result = client.decisions.create( model="gemini-decision-1", input=[{ "type": "image", "uri": "https://storage.googleapis.com/generativeai-downloads/images/spill.jpg", "mime_type": "image/jpeg", }, { "type": "text", "text": "Breakroom counter check before the 9 AM client meeting.", }, ], questions={ "has_spill": { "type": "noul", "instructions": "Is there an uncleaned liquid spill on the counter?", }, "next_action": { "type": "choice", "instructions": "Which cleanup action fits the scene?", "criteria": { "wipe_with_cloth": "Small counter spill with a cloth or towel already nearby.", "dispatch_janitorial": "Large floor hazard, broken glass, or active leak.", "ready_for_meeting": "Counter is clean and ready.", }, }, "severity": { "type": "score", "instructions": "Rate the severity of the counter mess.", "criteria": [ "Clean and ready.", "Minor localized spill.", "Major spill or safety hazard.", ], }, }, )
各团队如何基于 Decisions API 进行构建
Google 内部的多个团队已经在把 Decisions API 用作低延迟控制平面,只需定义选项就能交付功能,而不必训练定制分类器,也不用等待生成式调用返回:
Gemini Notebook 中的动态模型路由:Gemini Notebook 将 Decisions API 用作实时路由器,把进来的查询即时导向最适合该请求的模型层级与思考级别。Decisions API 可在毫秒级完成意图分类,让日常查询的延迟最多降低 40%,同时把深度推理模型留给复杂任务。
Stitch 中的快速设计迭代:Stitch 中新的快速画布实验使用 Decisions API,在数秒内生成产品流程的可交互线框图,并几乎即时地应用通过语音或点击做出的修改。当流程符合预期时,一键即可把屏幕移交给完整的 Stitch 体验,进行更深入的设计打磨。
CC 中的表情选择:来自 Google Labs 的 CC 使用 Decisions API,在几分之一秒内从 200 多个选项中挑选出合适的表情回应,因此表情回应可以即时出现,不会增加延迟。
Search 中 MCP 工具的安全层:Google Search 使用 Decision API,在延迟允许范围内高效地检测对抗性行为,并依据对抗性评估(adversarial evals)识别对抗性内容。
Stitch Rapid 使用 Decisions API,在数秒内设计出完整的产品流程(1 倍速)。
Gemini Decision 1 的定价为每 100 万输入 Token 0.10 美元,输出 Token 免费,并提供可在 Google AI Studio 中用于原型开发的 API 免费层。无论你评估多少个问题,共享输入每个请求只计费一次(每个问题只增加其自身的指令和标准 Token)。不收取输出 Token 费用,也没有可变的思考 Token 账单,或单独的音频转写和 OCR 费用。
你今天就可以开始构建:在 Google AI Studio 中获取 API key,试用我们的 Doorstep Parcels、Speed Sketch 和 Find the Zoo Keeper 演示应用,并按照 Decisions API 文档中的 Python、JavaScript 和 REST 指南进行操作。
生态集成正陆续登陆 OpenRouter、LangChain、Vercel、LiteLLM、Pydantic AI、Pipecat、LiveKit 和 OpenCode,同时还有 google-genai SDK,Google Cloud 中的 Gemini 也即将获得支持。
译文由英文原文机器翻译,术语按行业惯例统一,可在文末展开原文核对。
查看英文原文(Gemini Decisions API: Fast, calibrated multimodal decisions)
See all in Products & platforms
See all in Innovation & AI
See all in Products & platforms
Gemini Decisions API: Fast, calibrated multimodal decisions
Gemini Decisions API: Fast, calibrated multimodal decisions
Introducing the Gemini Decisions API, powered by our new Gemini Decision 1 model. Evaluate up to 1,000 yes/no, choice, and rubric-scoring questions in a fast, single pass across text, images, audio, video, and PDFs.
Product Manager, Google DeepMind
Principal Scientist, Google DeepMind
Your browser does not support the audio element.
We are introducing the Gemini Decisions API to help you make fast, probabilistic choices across text, images, audio, video, and PDFs. This tool computes confidence scores for multiple questions in a single pass, providing a cost-effective way to route queries, grade content, or check policy rules. You can start building today by accessing your API key in Google AI Studio and reviewing our documentation for Python, JavaScript, and REST guides.
Summaries were generated by Google AI. Generative AI is experimental.
Today, we’re introducing the Gemini Decisions API, a new endpoint for fast, closed-set decisions powered by our new Gemini Decision 1 model (gemini-decision-1), available through the Gemini API in Google AI Studio.
The Decisions API gives you purpose-built primitives for bounded choices like tool selection, policy checks, model routing, and context grading. You can pass one multimodal input across text, images, audio, video, and PDFs (up to 256K tokens) along with up to 1,000 independent questions. Rather than generating answer text token by token, Gemini Decision 1 scores your candidate options in a single forward pass, returning probabilities and confidence scores in 150-250 ms end-to-end latency at $0.10 per 1M input tokens with zero output tokens billed.

Strong performance and calibration
On benchmarks spanning agent routing, reasoning, ranking, and multimodal understanding, Gemini Decision 1 leads across the suite, with the lowest calibration error.
Gemini Decision 1 scores options without token-by-token decoding, keeping median latency 50-100 ms of model compute (150-250 ms end-to-end), across multimodal inputs and contexts up to 16K tokens, remaining nearly flat from 1 to 100 questions per request. With zero output tokens billed, Gemini Decision 1 with the Decisions API is nearly 5x cheaper across the shared benchmark suite than Gemini 3.5 Flash-Lite with the Interactions API.
Native multimodal evaluation across three decision types in one pass
Built on Gemini’s native multimodal architecture, Gemini Decision 1 can process text, images, audio, video, and PDFs directly in a single forward pass. You can combine modalities in one shared input, such as pairing a call recording with account metadata or a scanned PDF claim with damage photos, and tune resolution on image and video inputs to control visual detail and token cost.
You can evaluate multiple questions against that shared input in a single request, mixing yes/no checks, multiple choice, and rubric scores. Every answer includes a confidence score (from 0 to 1) so your code can route ambiguous cases to a fallback or human reviewer:
noul (Yes/No): Estimates the probability that a yes/no condition holds, with optional true and false description criteria.
choice (Multiple Choice): Selects from 2 to 255 choices, returning the winner and normalized probabilities for every candidate.
score (Ordered Rubric): Grades inputs across an ordered rubric (starting at 0), returning a continuous weighted score and per-level probabilities.
The input field uses the same Content block format as the Interactions API. Here is how to check a photo of a breakroom counter, together with a text note, against all three question types in one request:
from google import genai
client = genai.Client(http_options={"api_version": "v1beta"})
result = client.decisions.create(
model="gemini-decision-1",
input=[{
"type": "image",
"uri": "https://storage.googleapis.com/generativeai-downloads/images/spill.jpg",
"mime_type": "image/jpeg",
},
{
"type": "text",
"text": "Breakroom counter check before the 9 AM client meeting.",
},
],
questions={
"has_spill": {
"type": "noul",
"instructions": "Is there an uncleaned liquid spill on the counter?",
},
"next_action": {
"type": "choice",
"instructions": "Which cleanup action fits the scene?",
"criteria": {
"wipe_with_cloth": "Small counter spill with a cloth or towel already nearby.",
"dispatch_janitorial": "Large floor hazard, broken glass, or active leak.",
"ready_for_meeting": "Counter is clean and ready.",
},
},
"severity": {
"type": "score",
"instructions": "Rate the severity of the counter mess.",
"criteria": [
"Clean and ready.",
"Minor localized spill.",
"Major spill or safety hazard.",
],
},
},
)
How teams are building with the Decisions API
Teams across Google are already using the Decisions API as a low-latency control plane, shipping features simply by defining options rather than training custom classifiers or waiting on generative calls:
Dynamic model routing in Gemini Notebook: Gemini Notebook uses the Decisions API as a real-time router to instantly direct incoming queries to the optimal model tier and thinking level for each request. The Decisions API classifies intent in milliseconds to serve everyday queries with up to 40% lower latency while reserving deep reasoning models for complex tasks.
Rapid design iteration in Stitch: The new rapid canvas experiment in Stitch uses the Decisions API to generate playable wireframes of a product flow in seconds and to apply edits made by voice or tap almost instantly. When the flow feels right, one click hands the screens off to the full Stitch experience for deeper design polish.
Emoji selection in CC: CC from Google Labs uses the Decisions API to pick the right emoji reaction from a palette of more than 200 options in a fraction of a second, so reactions appear instantly without adding latency.
Safety layer for MCP tools in Search: Google Search uses the Decision API to detect adversarial behavior and content against adversarial evals with high efficacy within the latency envelope.
Stitch Rapid uses the Decisions API to design an entire product flow in seconds (1x speed).
Gemini Decision 1 is priced at $0.10 per 1M input tokens and $0 for output tokens, with an API no-cost tier you can use in Google AI Studio for prototyping. Your shared input is billed once per request no matter how many questions you evaluate (each question adds only its own instruction and criteria tokens). There are no output-token charges, variable thinking-token bills, or separate audio transcription and OCR costs.
You can start building today by grabbing an API key in Google AI Studio, trying our Doorstep Parcels, Speed Sketch, and Find the Zoo Keeper demo apps, and following the Python, JavaScript, and REST guides in the Decisions API documentation.
Ecosystem integrations are rolling out across OpenRouter, LangChain, Vercel, LiteLLM, Pydantic AI, Pipecat, LiveKit, and OpenCode alongside the google-genai SDKs, with support coming soon to Gemini in Google Cloud.
原文为英文,正文由自动抓取并保留原文。
原文发布于 Google Blog(Gemini Decisions API: Fast, calibrated multimodal decisions),标题、摘要与正文由机器翻译,术语按行业惯例统一,未经人工逐句校订。