Skip to content

model

DeepSeek V4.1-Flash

Multimodal mixture-of-experts model from DeepSeek built around long-context serving, with a 552B-parameter backbone, a separate memory subsystem and support for million-token inputs.

Known aliases

  • DeepSeek V4.1 Flash
  • V4.1-Flash

Relationships

No evidence-backed relationships are recorded.

Current clusters