V4 Preview: Entering the Era of Affordable Million

thinking in tool-use， more efficient training inference on long context， faster inference， with native visual understanding. Designed for greater capability， and higher throughput. Read more Introducing DeepSeek-V3.2-Exp DeepSeek-V3.2-Exp debuts DeepSeek Sparse Attention (DSA) for faster， CMO， faster，。

DeepSeek-V3.2: Pushing the Frontier of Open Large Language Models DeepSeek-V3.2 and V3.2-Speciale launch with world-leading reasoning， ICPC IOI 2025. Read more Introducing DeepSeek-V4.1-Flash: smarter， more efficient. Introducing the smallest model in our new architecture family。

with API prices cut by 50%+. ， and gold-medal performance in IMO。