DeepSeek API platform recently announced the resumption of its API recharge service, which was temporarily suspended on February 6th due to server resource constraints.
Along with the restoration of the recharge function, DeepSeek has also adjusted the pricing for some model calls. The promotional period for the deepseek-chat model has ended. The pricing for its underlying V3 model is now ¥2 per million input tokens and ¥8 per million output tokens. The deepseek-reasoner model, based on the R1 model, is priced at ¥4 per million input tokens and ¥16 per million output tokens.
According to DeepSeek's official explanation, a token is the smallest unit used by the model to represent natural language text, which can be words, numbers, or punctuation marks. DeepSeek will calculate and charge based on the total number of input and output tokens. The deduction will be the token consumption multiplied by the model's unit price. If a user has both a recharge balance and a gift balance, the system will prioritize deducting from the gift balance.
The resumption of DeepSeek's API recharge service and the price adjustments will undoubtedly impact users. Users should pay attention to the new pricing and plan their recharges and usage accordingly.