What are the core components of Pixtral-12B-2409?

Question

Answers ( 1 )

    0
    2025-04-01T05:32:47+00:00

    Pixtral-12B-2409 consists of two primary components:
    - **Decoder**: A 12-billion-parameter transformer model for text processing.
    - **Visual Encoder**: A 400-million-parameter module for image understanding.
    The model is designed to handle interleaved image-text data natively.

Leave an answer