DeepSeek makes its 75% discount on V4 Pro permanent, cutting prices fourfold

DeepSeek has announced that its 75% discount on its flagship V4 Pro model will continue indefinitely. Last week, the promotion was due to end on May 31, but the company decided to make it permanent.

The numbers are startling

DeepSeek's new prices look almost unbelievable. Input tokens now cost $0.003625 per million, and output tokens cost $0.087. Before, they were four times higher: $0.0145 and $0.35 respectively.

By comparison, OpenAI's GPT-5 costs $2.50 per million input tokens, which is 689 times more. Even the budget models from OpenAI and Anthropic remain noticeably more expensive than DeepSeek. Claude 3.5 Sonnet costs $3 for input tokens. Grok sits in a similar price range to OpenAI.

The price war has reached the core

DeepSeek's decision to keep the discount is not just a PR move. It signals that the Chinese startup intends to stay aggressively cheaper than all its competitors. The offer was originally announced as limited, but the extension suggests that competition is forcing the company's hand.

Here is what happened:

• V4 Pro has reached top positions in global benchmarks
• DeepSeek is now 4-8 times cheaper than the market leaders
• Companies are starting to review their contracts for OpenAI models
• Competitors are being forced to cut prices or lose customers

A new phase of competition

The AI model market has entered a new phase: it is no longer a war of features but a war of prices. DeepSeek shows that quality models can be built cheaply, and profitably. Large companies that still overpay for GPT-4 or GPT-5 will start counting every penny.

What this means

AI prices will fall even further. If OpenAI, Anthropic and other leaders do not respond aggressively to this move, they will lose some customers by autumn. At the same time, DeepSeek's low price may become a barrier to adoption in the enterprise sector, since companies usually choose proven solutions rather than deciding on price alone.