Google recently updated the Gemini API pricing structure, introducing five service tiers: Standard, Flexible, Priority, Batch, and Cache. The Flexible and Batch tiers offer a 50% discount on standard rates, respectively suited for scenarios with low latency sensitivity (1–15 minutes) and ultra-large-scale data processing (up to 24-hour latency). The Cache tier is billed based on the number of tokens and storage duration, ideal for high-frequency, complex instruction calls. The Priority tier carries a 75%–100% premium, ensuring millisecond-to-second response times for critical applications such as customer service bots and real-time fraud detection. This adjustment enhances resource allocation capabilities for AI inference services, providing a more granular pricing model for AI applications with varying latency sensitivity and cost constraints.
Author and source: AIBase
Google has recently updated the billing structure for its Gemini API to better meet users' inference needs. This update introduces several new service tiers, including Standard, Flexible, Priority, Batch, and Cached. Users can choose the most suitable tier based on their specific requirements.
First, the Standard tier provides basic inference services, allowing users to choose based on their usage needs. The Elastic tier is an innovative option that leverages idle computing resources during off-peak hours, offering users a 50% discount off the standard price. This tier targets a latency range of 1 to 15 minutes but does not guarantee fixed latency, making it ideal for applications with less stringent time requirements.
In addition, the bulk tier offers users a 50% discount on standard fees, making it ideal for those handling large volumes of data, with a maximum latency of up to 24 hours. This tier is particularly suited for large-scale data processing scenarios, allowing users to significantly reduce costs when performing extensive information queries.
For the cached tier, billing is based on the number of cached tokens and the duration of storage, making it ideal for chatbots requiring frequent invocation of complex commands, long-form video analysis, or queries on large document sets. This tier enables users to efficiently manage storage and computational resources, enhancing system performance.
The priority tier is priced 75% to 100% higher than the standard rate but enables latency control in milliseconds to seconds. This tier is ideal for applications requiring real-time responses, such as customer service chatbots, real-time fraud detection, and mission-critical business assistants. Google recommends users with such needs select the priority tier to ensure optimal speed and efficiency for their applications.
Key points:
🌟 Added multiple Gemini API service tiers to meet the needs of different users.
⏳ Flexible and batch tiers offer a 50% discount, ideal for large-scale data processing.
⚡ Priority tier ensures millisecond-level response, ideal for real-time applications.