DeepSeek Makes Aggressive 75% Price Cut to Weaponize Western AI Rate Limit Anger

Although it’s been somewhat quiet lately, DeepSeek remains a major player in the tech industry. The company made headlines some time ago for its powerful capabilities and affordable prices. Now, bucking the trend, the company is taking things a step further. DeepSeek has officially made its massive 75% price cut on its flagship DeepSeek-V4-Pro API permanent, offering higher AI usage limits for even less money.

DeepSeek locks in Permanent 75% discount for V4-Pro AI model

Initially, DeepSeek framed the discount as a short-term promotional campaign in April. However, given the user backlash against increasingly restrictive Western AI chatbots, the company has made the massive shift. This way, DeepSeek wants to attract a large wave of developers and enterprise users dissatisfied with its competitors.

Over the last few weeks, developer forums and tech communities have filled with growing user frustration. Subscribers to prominent Western consumer chatbots and professional APIs have reported hitting sudden, restrictive usage caps. The list mainly includes Google Gemini, Anthropic’s Claude, and even Perplexity. Because these platforms struggle to manage the immense infrastructure costs of advanced computing, they have quietly tightened user gateways through complex “compute-used” monitoring, resulting in frequent service lockouts.

DeepSeek is moving quickly to weaponize this friction. While Western ecosystems restrict availability to protect their servers, DeepSeek is making its rock-bottom pricing permanent to offer an unrestricted, long-term haven. With these permanent, heavily discounted rates, the platform provides a highly stable alternative for creators who want to deploy complex autonomous agents or large-scale code bases without fearing a sudden digital brick wall.

The rock-bottom math of the permanent V4-Pro tier

Making the 75% discount permanent locks in DeepSeek V4 Pro rates that are a mere fraction of what U.S. competitors demand. Non-cached input tokens on the model will permanently remain at a microscopic $0.435 per million tokens. This is down from the original $1.74 base price. Output fees are also permanently anchored at just $0.87 per million tokens, a deep dive from the previous $3.48 mark.

When measured against the competition, running identical data-heavy operations on frontier American models like OpenAI’s flagship series or top-tier Claude variations can easily cost between 8 and 35 times more.

Furthermore, DeepSeek cemented this advantage by pairing the permanent V4-Pro drop with a definitive 90% cost reduction for input cache hits across its entire API line. This means that for repetitive prompts or continuous system instructions, developers are charged a nearly negligible fee, making iterative software testing incredibly cheap over long development cycles.

A structural play for market dominance

Industry observers note that sacrificing immediate profit margins aligns perfectly with DeepSeek’s broader ambitions as it pursues a massive $45 billion funding round. The firm previously shook up the hardware ecosystem by demonstrating that it could deliver top-tier intelligence with drastically optimized operational overhead.

DeepSeek’s move is, all in, to ride the wave of maximum customer resentment against Western subscription restrictions and to persuade disappointed power users that they have more than a temporary alternative but a permanent new home for unrestricted AI development.

The post DeepSeek Makes Aggressive 75% Price Cut to Weaponize Western AI Rate Limit Anger appeared first on Android Headlines.

​