DeepSeek

Hangzhou, ChinaDal 2023deepseek.com

In inglese

DeepSeek publishes the weights of its current models under the MIT licence. The largest, DeepSeek-V4-Pro, has 1.6T parameters.

Di punta

DeepSeek-
V4.1-
Flash

Di puntaIn inglese

A 552B mixture of experts built as a causal encoder and decoder, with 8B parameters active per token in prefill and 16B in decode.

Leggi: DeepSeek-V4.1-Flash

Gamma

Misura
  1. DeepSeek-V4.1-Flash

    Di punta

    Totali
    552B
    Attivi
    16B
    Contesto
    1M
  2. DeepSeek-V4-Pro

    Totali
    1,6T
    Attivi
    49B
    Contesto
    1M

Altri provider