DeepSeek permanently cuts V4-Pro AI model prices by 75%

DeepSeek will keep the price of its flagship V4-Pro artificial intelligence model at a quarter of its original level, after announcing a permanent 75% cut on Saturday.

The Chinese startup did not say whether the move was tied to a larger supply of Huawei’s Ascend 950 chips, which it used to maximize V4’s performance.

API pricing for V4-Pro

The company said V4-Pro API costs now range from 0.025 to 6 yuan per million tokens, or about $0.0035 to $0.83, depending on usage type. That compares with the previous range of 0.1 to 24 yuan per million tokens.

A token is a unit of text processed by the AI model.

Huawei chips and supply limits

Huawei’s AI chip sales have benefited from U.S. export controls that stop Nvidia from selling its most advanced semiconductors in China. Separate restrictions on chipmaking equipment exports, however, have limited Huawei’s ability to scale up Ascend production.

When DeepSeek introduced V4 last month, it said the Pro version would cost up to 12 times more than the less powerful Flash version because of “constraints in high-end compute capacity,” which limited availability.

The company also said Pro pricing was expected to fall sharply once Huawei Ascend 950 supernodes are launched in large quantities in the second half of the year.