DeepSeek has permanently reduced the price of its flagship V4 model by 75 percent, aiming to undercut competition in the AI agent market. The company is positioning itself as the “cost-effective” choice for AI agents following the price adjustment, which sets the DeepSeek V4 Pro at a quarter of its initial cost.
According to the company’s website, new prices for the DeepSeek V4 Pro now range from $0.003625 to $0.87 per one million tokens, down from a previous range of $0.0145 to $3.48 per million tokens. This price change makes permanent a promotion that was previously set to expire on May 31, 2026.
The decision to lower prices comes one month after the release of DeepSeek’s V4 models, Pro and Flash. DeepSeek claims these models will usher in an “era of cost-effective 1M context length.” The significant price reduction is anticipated to provide major cost savings for enterprise accounts and power users who require extensive token usage.
Furthermore, the price cut offers a more affordable option compared to leading AI models, such as OpenAI’s GPT-5 and Google’s Gemini 3.5 Flash. This aggressive pricing strategy may provoke competitors like Anthropic, which has previously accused DeepSeek of “distillation attacks” on its AI models.
