DeepSeek Full-Series API Input Cache Price Cut to 1/10 of Original, Plus Limited-Time Discounts as Low as 0.02 Yuan per Million Tokens

Starting today, the input cache hit price for DeepSeek's full-series API services has been reduced to 1/10 of the original price. This adjustment covers all models, allowing users to save more costs when using the API.

Additionally, the Pro model has a limited-time 25% discount until May 5, 2026. With the combined discount, DeepSeek-V4-Pro model input (cache hit) is as low as 0.025 yuan per million tokens, and DeepSeek-V4-Flash is 0.02 yuan per million tokens, further lowering the threshold for using AI services.