DeepSeek cuts V4-Pro prices by 75% (thenextweb.com)

DeepSeek has announced a dramatic 75% discount on its newly released V4-Pro AI model, effective until May 5, 2026. This pricing strategy not only positions V4-Pro as a low-cost alternative to competitors like OpenAI’s GPT-5.5 and Google's Gemini 3.1 Pro but also intensifies the ongoing pricing competition among AI providers. Even before the discount, V4-Pro's token costs undercut major players, with the promotional price bringing input costs down to approximately $0.036 per million tokens. The strategic cut in API prices, particularly for cache hits, is aimed at frequent users and enterprises, streamlining adoption by reducing cost barriers for large-scale deployments. The significance of this move unfolds in the geopolitical context of the US-China AI landscape, where the US has accused Chinese firms of trying to replicate American AI models. DeepSeek's pricing strategy can be seen as both a competitive challenge and a political statement amid these tensions. With its 1.6 trillion parameter mixture-of-experts model optimized for non-Nvidia hardware, V4-Pro marks a potential shift in the AI landscape, enhancing the viability of domestic AI solutions and pushing longer-context processing into mainstream applications. As industry analysts suggest, this approach could accelerate the global AI development pace, particularly in jurisdictions less reliant on Nvidia’s technology.