Gemini Decisions API
A multimodal decision API introduced by Google, powered by a specially optimized Gemini Decision model. It provides fast, confidence-calibrated judgment outputs and is suited to scenarios such as risk control, content moderation, and automated workflows.
Core Facts
Timeline (last 90 days)
The release of Gemini 4 Hydrogen was accompanied by two actions: the Gemini Decisions API launching in AI Studio, and smart glasses opening for pre-order.
Gemini Decision 1, the underlying model of the Gemini Decisions API, supports uploading the same material only once to answer 1,000 independent questions; the interface bills by input, charges no output token fees, and input costs about $0.1 per million tokens.
The Gemini Decisions API charges only for input tokens and has not disclosed how output is billed or any additional fees.
The closed-ended question design of the Gemini Decisions API requires manual pre-definition and is not suitable for open-ended, exploratory tasks.
The article does not provide benchmark data for the Gemini Decisions API, so its reliability in professional domains cannot be judged.
The article does not explain the specific format requirements and size limits for multimodal input to the Gemini Decisions API.
The Gemini Decisions API wraps large model capabilities into a narrower, more controllable interface, reducing integration complexity and shortening the path from prototype to production.
The Gemini Decisions API supports up to 1000 yes/no, choice or rating-criteria questions per request.
The Gemini Decisions API returns calibrated probability results and does not output free-form answers.
Gemini Decision 1 supports text, image, audio, video, and PDF document inputs, and outputs boolean probabilities, choices, and scores, without consuming output tokens.
3 more timeline events
All Facts (13)
Google's Gemini Decision 1 model launched on AI Gateway, positioned as a decision model rather than a content generation model.
70%UnverifiedGemini Decision 1, the underlying model of the Gemini Decisions API, supports uploading the same material only once to answer 1,000 independent questions; the interface bills by input, charges no output token fees, and input costs about $0.1 per million tokens.
50%UnverifiedThe Gemini Decisions API wraps large model capabilities into a narrower, more controllable interface, reducing integration complexity and shortening the path from prototype to production.
50%UnverifiedThe article does not provide benchmark data for the Gemini Decisions API, so its reliability in professional domains cannot be judged.
50%UnverifiedThe article does not explain the specific format requirements and size limits for multimodal input to the Gemini Decisions API.
50%UnverifiedThe Gemini Decisions API supports up to 1000 yes/no, choice or rating-criteria questions per request.
50%UnverifiedThe Gemini Decisions API returns calibrated probability results and does not output free-form answers.
50%UnverifiedThe Gemini Decisions API charges only for input tokens and has not disclosed how output is billed or any additional fees.
50%UnverifiedThe closed-ended question design of the Gemini Decisions API requires manual pre-definition and is not suitable for open-ended, exploratory tasks.
50%UnverifiedGemini Decision 1 supports text, image, audio, video, and PDF document inputs, and outputs boolean probabilities, choices, and scores, without consuming output tokens.
50%UnverifiedThe Gemini Decisions API targets invocation scenarios that require low-latency return of judgment results.
50%UnverifiedThe release of Gemini 4 Hydrogen was accompanied by two actions: the Gemini Decisions API launching in AI Studio, and smart glasses opening for pre-order.
50%UnverifiedThe Gemini Decisions API is driven by the specially optimized Gemini Decision model and is positioned to provide fast, calibrated multimodal decision-making capabilities.
50%