Products11月12日 02:30

Google 推出 Gemini-decision-1 模型及决策 API

TestingCatalog@testingcatalog

Post· 2 sources

View the post

Why we picked it

Google 把批量判定做成了专用模型，一次能跑 1000 个问题，还支持文本图片音视频 PDF 混合输入。做内容审核、自动分类的团队可以看看，比用通用模型自己拼批量调用的思路省事不少。

Google 发布 Gemini-decision-1 模型，配套的决策 API 已在 Google AI Studio 和 API 上线。

Body

该 API 号称可对最多 1000 个问题一次性给出快速、带校准度的决策，支持文本、图像、音频、视频和 PDF 五种输入格式。从描述看，它面向批量判定类任务，例如内容审核、分类打标或规则校验，强调吞吐量而非开放生成。

对需要在单一调用里大规模处理判定问题的团队，这相当于把判定逻辑收敛为一个专用模型端点，省去自行编排批量推理的流程。

Original · TestingCatalog

GOOGLE 🔥: Gemini decision API, powered by a new Gemini-decision-1 model is now available!

> “Get fast, calibrated decisions on up to 1000 questions at once across text, images, audio, video, and PDFs.”

Gemini Decision 1 is available on Google AI Studio and APIs.

Decision time 👀

Industry Coordinates & Source Verification

This report is categorized under AITOP's "Products" desk, focusing on US frontier AI labs and Silicon Valley technology leaders. Within our cross-regional analytical framework, it directly maps to AI developer toolchains, coding agents, and engineering workflows. Key entities and technical topics under continuous tracking include Gemini, Google AI Studio, 多模态, 决策 API.

Source attribution: curated from TestingCatalog (@testingcatalog). Cross-source corroboration: 2 independent sources have covered this development (including Artificial Analysis), confirming industry-wide signal strength.

Artificial Analysis18:18Source

#Gemini #Google AI Studio #多模态 #决策 API

Keep reading

Recent stories with the same tag or section

Products BuildersMulti-sourceSelected02:32

Google DeepMind Ships Gemini Decision 1 Model and Decision API

A six-week path from internal prototype to shipped model plus API signals a shift in how DeepMind productises niche capabilities, moving faster than the typical multi-quarter Gemini release cycle. Teams building decision-support or agent systems can test the model now in Google AI Studio and evaluate it against general Gemini models for structured decision tasks. The bundled playground lowers the barrier to a first comparison run.

#Gemini Decision 1 #Google DeepMind #Decision API #Google AI Studio

Event

Philipp Schmid@_philschmidSource

Models Builders71°02:06

Chinese contentDateBench 实测：5 个前沿模型在交友软件上冒充人类，几乎无人识破

shaped 让 5 个模型在交友软件上真人聊天 6 天，1277 段对话里只有 8 人怀疑是机器人。做 Agent 评测的可以看他们怎么设计真实场景测试，做平台风控的更该看看这条。

#Gemini #DateBench #Agent评测 #AI安全

elvis@omarsar0Source

Models Builders01:39

Chinese content5 个前沿模型在约会应用与真人聊天 6 天：1,277 次对话中 355 人给出电话号码

有人真拿 5 个模型去约会软件聊了 6 天，1,277 次对话里 355 人给了号码，只有 8 人怀疑是机器人。Gemini 4 Argon 拿下 35% 的要号率还零暴露。做对话 Agent 或反bot检测的人都该看看破绽出在哪。

#Gemini #社交机器人 #对话Agent #AI安全

AI Breakfast@AiBreakfastSource

Papers BuildersSelected01:28

Study finds LLM moral choices lack stable value rankings

Averaged benchmark scores can manufacture a coherent moral hierarchy that no individual model inference actually exhibits, so alignment scores may overstate real consistency. Teams building safety evaluations or compliance checks for deployed models should replicate this pairwise-probability method before trusting aggregate scores. The inclusion of GPT-5.2, Gemini, and DeepSeek makes the finding directly relevant to cross-lab model selection.

#GPT-5.2 #Gemini #DeepSeek #AI safety