DeepSeek to raise API prices for V4 models by up to 1100%

Chinese artificial intelligence startup DeepSeek is set to implement a significant increase in its API pricing for its V4-Pro and V4-Flash models, alongside the introduction of a peak and off-peak pricing structure. These changes, detailed in a company statement, will take effect on August 17. The new rates represent a substantial jump, ranging from 50% to 1,100% above current prices, with the exact increase depending on the specific model, token type, and the time of usage. The company also formally released its V4 Pro model, which it states greatly enhances agent capabilities and is available through API, APP, and Web channels. DeepSeek's V4 models support a context length of 1 million tokens and have a maximum output of 384,000 tokens. The company's strategy has been to offer competitive pricing, with its V4 Pro model priced below Western mid-tier options and its V4 Flash model priced very low. The upcoming changes aim to adjust these rates while maintaining a competitive position in the market. DeepSeek V4-Pro-0813 is priced at $1.32 per million input tokens and $3.96 per million output tokens.