Tech News
← Home  ·  All topics

Compressed Kv Caching

1 GoKawiil brief on this topic

DeepSeek releases V4.1 Flash 1017, its first CED-architecture model

DeepSeek has shipped V4.1 Flash 1017, dated October 17, 2026, as a snapshot of its V4.1 Flash model available via the deepseek-flash API. The sparse mixture-of-experts model uses a new Causal Encoder-Decoder architecture, activating 8B parameters on input and 16B on output from a 552B-parameter backbone, with native image understanding trained jointly with text from the start.