DeepSeek Unveils Preview of Highly Anticipated V4 Large Language Model in Major AI Breakthrough

Chinese artificial intelligence startup DeepSeek has released a preview version of its long-awaited V4 large language model, marking a significant development in the global AI competition as of April 24, 2026.

The Hangzhou-based company announced the release of the V4 model preview on Friday, allowing users to test its new capabilities and features. This follows more than a year after DeepSeek introduced its R1 reasoning model, which disrupted global tech markets due to its strong performance at a lower cost compared to models from companies like OpenAI and Google.

DeepSeek claims that the V4 model achieves strong performance against domestic competitors, particularly in agent-based tasks, knowledge processing, and inference. The company has optimized the model for use with popular agent tools such as Anthropic’s Claude Code and OpenClaw.

The V4 model is available in both a “pro” and a “flash” version, depending on size. Like its predecessors, the V4 model is open source, enabling developers to download the code, run it locally, and modify it according to their needs.

DeepSeek gained attention in late 2024 with its free, open-source V3 model, which the company said was trained using less powerful chips and at a fraction of the cost of models built by larger AI firms. The subsequent release of the R1 reasoning model in January 2025 further solidified its reputation for delivering competitive performance while maintaining cost efficiency.

The release of the V4 preview intensifies the technological competition between the United States and China in the artificial intelligence sector. According to reports, DeepSeek has not shared its upcoming AI model with American engineers and has instead granted early access to Chinese companies, underscoring the growing strategic divide in AI development between the two nations.

Founded in 2023, DeepSeek continues to position itself as a key player in the global AI landscape through its commitment to open-source models and cost-effective innovation. The company’s latest move with the V4 preview reflects its ongoing strategy to challenge established AI leaders by offering high-performing alternatives that are accessible to a broader range of developers and organizations.