DeepSeek API Implements Peak and Off-Peak Pricing
IT Home reports that DeepSeek API's new pricing structure, differentiating between peak and off-peak hours, has gone into effect. Prices will double during peak usage times.
IT Home, a Chinese technology publication, announced on August 17 that DeepSeek API has implemented a new pricing model that adjusts rates based on usage times. This new structure, effective August 17, 2026, introduces higher prices during periods of peak demand.
The new scheme doubles API call costs during peak hours, defined as 9:00 AM to 12:00 PM and 2:00 PM to 6:00 PM Beijing time. All other times are considered off-peak.
This pricing adjustment affects the deepseek-v4-flash and deepseek-v4-pro models. For instance, the deepseek-v4-flash model's input cost (cache hit) during peak hours is 0.10 yuan per million tokens, compared to 0.05 yuan during off-peak hours. Input costs (cache miss) rise to 3.0 yuan from 1.5 yuan, and output costs increase to 9.0 yuan from 4.5 yuan during peak times.
Previously, the DeepSeek V4 Pro model had a different pricing structure, with input costs (cache hit) at 0.025 yuan per million tokens, input costs (cache miss) at 3 yuan, and output costs at 6 yuan.
The DeepSeek V4 models are noted for their extended context capabilities and have been recognized domestically and within the open-source community for their performance in agent functions, world knowledge, and reasoning. The model is available in two versions based on size.