DeepSeek Offers Steep Discounts for Off-Peak AI Access
Hangzhou-based Chinese artificial intelligence (AI) company DeepSeek has introduced a new discount program to address the strain on its server resources caused by high demand. The company is slashing prices for access to its models via its application programming interface (API) during off-peak hours.
DeepSeek is cutting the price of accessing its models through its API during off-peak hours after midnight. Photo: Reuters
The new rates, announced on Wednesday, will take effect after midnight on Thursday. From 12:30 a.m. to 8:30 a.m. China time, API access to the V3 model will be available with a 50 percent discount. This translates to US$0.035 per million tokens for cache hits, US$0.135 per million tokens for cache misses, and US$0.55 per million tokens for output.
Access to DeepSeek’s R1 reasoning model will be available during the same hours with a 75 percent discount. During off-peak hours, the two models are priced the same.
Context length—the maximum number of tokens a model can process at once—is 64,000 tokens for both models. In the context of AI, a token represents a fundamental unit of data processed by the algorithm, which can include a word, number, or even punctuation. DeepSeek charges users based on the total number of input and output tokens processed by its models.
For comparison, OpenAI’s o1 reasoning model is priced at US$15 per million input tokens, US$7.5 per million cached input tokens, and US$60 per million output tokens, with a context length of 200,000 tokens.
