[Google Updates Gemini API Pricing: Tiered Billing Based on Inference Usage] Google recently updated the pricing tiers for the Gemini API, with optimized plans and pricing designed based on actual inference usage needs. The newly added inference service tiers include: Standard, Flex, Priority, Batch, and Caching versions. The Flex inference tier offers a 50% discount on standard pricing by utilizing off-peak idle computing resources, with a target latency of 1 to 15 minutes, though no latency guarantees are provided. Similarly, the Batch API tier also offers a 50% discount on standard rates, with a maximum latency of up to 24 hours. The Caching tier charges based on the number of cached tokens and storage duration, and is recommended for scenarios such as dialogue bots equipped with complex system instructions, repeated analysis of long video files, and large-scale document set queries. The Priority tier is priced 75% to 100% higher than the standard rate, with latency controlled within milliseconds to seconds. Google recommends this tier for use cases such as real-time customer service chatbots, real-time fraud detection, and mission-critical intelligent assistants.