Google Launches Gemini Decisions API With Gemini Decision 1
Google debuts Gemini Decisions API powered by Gemini Decision 1, delivering calibrated probabilities for up to 1,000 questions at $0.10 per million tokens.
Stock photo for illustration only, not from the actual event
- Google launches Gemini Decisions API powered by the new Gemini Decision 1 model
- Returns calibrated probabilities for up to 1,000 yes/no, choice, and rubric questions
- Supports multimodal inputs including text, images, audio, video, and PDF files
- Priced cost-effectively at $0.10 per 1 million input tokens
Google has officially introduced the Gemini Decisions API, powered by its advanced Gemini Decision 1 model. This new developer tool is designed to process complex analytical queries and deliver precise probability calculations across large datasets.
A core feature of the Gemini Decisions API is its ability to return finely calibrated probabilities for a wide variety of inquiries. It supports up to 1,000 yes/no questions, multiple-choice selections, and rubric-based evaluation questions within a single workflow.
Stock photo for illustration only, not from the actual event
Furthermore, the system demonstrates robust multimodal capabilities, accepting diverse input formats such as text, images, audio, video, and PDF documents. This allows users to analyze complex, cross-format information seamlessly and efficiently.
The launch of the Gemini Decisions API highlights Google's ongoing strategy to make enterprise-grade artificial intelligence more accessible and affordable for quantitative analysis. By handling thousands of probability evaluations at a rate of $0.10 per million tokens, the API significantly lowers the financial barrier for organizations implementing large-scale automated decision-making systems.
Regarding pricing structure, Google has set the rate for the Gemini Decisions API at $0.10 per 1 million input tokens, positioning it as an economical solution for developers and enterprises looking to integrate advanced decision-support capabilities into their applications.
Source: MarkTechPost
Found something wrong in this article? Report an issue with this article
Comments
Leave a Comment