DeepSeek to quadruple API prices as new V4 models arrive

DeepSeek, known for undercutting Western AI rivals on price, is preparing customers for a major shift. With the launch of its V4 Pro model, the company will raise API pricing fourfold, introducing peak and off-peak rates it says will help "allocate resources more reasonably."

Starting August 16, V4 Pro will cost $3.96 per million output tokens during peak hours - more than four times the current $0.87 rate. Off-peak usage will be half that, at $1.98. The lighter V4 Flash model jumps from $0.28 to $1.32 per million output tokens at peak, dropping to $0.66 off-peak.

The company had earlier marked current pricing as a promotional deal ending May 31, then announced in May that the discounts would become permanent. It has since reversed course and opted for the increases.

Even after the hike, DeepSeek remains cheaper than some rivals. Moonshot's Kimi K3 costs $15 per million output tokens, and OpenAI's top-tier GPT-5.6 Sol runs $30. However, OpenAI's budget option, GPT-5.6 Luna, is priced at $1.20 - under DeepSeek's V4 Flash peak rate.

Source: Engadget