DeepSeek Unveils V4 Models With 1 Million Token Context Window, Challenging OpenAI and Anthropic
China's DeepSeek released preview versions of its V4 Flash and V4 Pro models, claiming top-tier performance in coding benchmarks and major advancements in reasoning and agentic tasks. The new models support a 1 million-token context window and introduce a Hybrid Attention Architecture for improved long-conversation memory.
DeepSeek V4 Announcement
A year after rattling Silicon Valley with its technology, China's DeepSeek rolled out preview versions of a new flagship artificial intelligence model, calling it the most powerful open-source platform in a challenge to rivals from OpenAI to Anthropic PBC. The Chinese startup unveiled the V4 Flash and V4 Pro series, touting top-tier performance in coding benchmarks and big advancements in reasoning and agentic tasks.
Technical Innovations
DeepSeek singled out a technique it dubbed Hybrid Attention Architecture, which it said improves the ability of an AI platform to remember queries across long conversations. It also pushed the 1 million-token context window — a leap that allows entire codebases or long documents to be sent as a single prompt.
Competitive Standing
Chinese startup says DeepSeek-V4-Pro beats all rival open models for maths and coding. While Silicon Valley retains a slight edge in developing the most advanced models, Chinese companies have "effectively closed" the AI performance gap with their US rivals, according to the Stanford AI Index 2026.