Gemini Decision 1
Google released Gemini Decision 1 and the Gemini Decisions API, which scores up to 1,000 closed-set questions per request in one forward pass at $0.10 per 1M input tokens.
Gemini Decision 1 (gemini-decision-1) scores candidate options in a single forward pass instead of generating answer text token by token. It returns probabilities and confidence scores in 150-250 ms end-to-end, and zero output tokens are billed. One multimodal input of up to 256K tokens (text, images, audio, video, PDFs) can be paired with up to 1,000 independent yes/no, choice and rubric-scoring questions.
- Date
- Wednesday 11 November 2026
- Lab
- Kind
- model
- Access
- closed API
- Price
- $0.10 per 1M input tokens, zero output tokens billed
Figures
| Measure | Value | Measured by |
|---|---|---|
| End-to-end latency | 150-250 ms 50-100 ms of model compute | company |
| Input price | $0.10 per 1M input tokens zero output tokens billed | company |
| Max questions per request | 1,000 | company |
| Max input context | 256K tokens | company |
Available through the Gemini API in Google AI Studio. A social post excerpted on Techmeme compares it with a rival model.
Sources
- blog.google/innovation-and-ai/technology/developers-tools/gemini-decisions-api/
- techmeme.com/261111/p40#a261111p40
- paidfreedroid.com/2026/11/google-gemini-decision-1-model-decisions-api-launch.html
- techiexpert.com/google-launches-gemini-decision-1-a-fast-ai-model-built-for-making-decisio
- latent.space/p/ainews-d-day-decision-models-day/
- eu.36kr.com/en/p/4066443558491008
This record was checked against its sources on 11 November 2026. How we check