Google announces Gemini Decisions API for one-pass multimodal decisions
🎧 Listen to this story
Google announced the Gemini Decisions API, available through the Gemini API in Google AI Studio. The company says the endpoint uses Gemini Decision 1 to evaluate up to 1,000 independent yes/no, multiple-choice and rubric-scoring questions against the same multimodal input, including text, images, audio, video and PDFs of up to 256K tokens. Responses include probabilities and confidence scores from 0 to 1.
Google says the model returns results in a single pass, with 150–250 milliseconds of end-to-end latency, a cost of $0.10 per 1 million input tokens and no output-token charges. The company also says Gemini Decision 1 had the best calibration performance across its benchmarks and was nearly five times cheaper than Gemini 3.5 Flash-Lite with the Interactions API on the shared benchmark suite.
Why it matters · editorial interpretation
For teams building multimodal evaluation workflows, the API could reduce repeated calls by supporting up to 1,000 decisions on the same input in a single pass. Google’s claims about latency, cost and calibration suggest a more economical option for tasks requiring many structured checks.
Sources
APIs