DeepSeek announced today: “We plan to raise the pricing of DeepSeek API services across the board in the near future, with a relatively large increase expected. Please plan your usage accordingly. The specific plan will be subject to the official announcement.”
Currently, DeepSeek API pricing is based on “input tokens” and “output tokens,” which are billed separately. Prices also vary depending on whether the cache is hit or missed. The main model prices are as follows:
1. DeepSeek-V4-Flash (preferred for high cost-effectiveness / high-concurrency scenarios)
Input (cache hit): 0.02 yuan / million tokens
Input (cache miss): 1 yuan / million tokens
Output: 2 yuan / million tokens
2. DeepSeek-V4-Pro (preferred for high-performance / complex reasoning scenarios)
Input (cache hit): 0.025 yuan / million tokens
Input (cache miss): 3 yuan / million tokens
Output: 6 yuan / million tokens
