DeepSeek publishes its V4 model series under an MIT license

DeepSeek published a preview of its V4 series on 26 April 2026, comprising V4-Pro at 1.6 trillion total parameters with 49 billion active and V4-Flash at 284 billion with 13 billion active, released under the MIT license with checkpoints on Hugging Face.

Why it mattered The accompanying paper reported that, against DeepSeek-V3.2, the new models needed 27% of the inference compute and 10% of the memory cache at million-token context lengths.