• boissez@alien.topB
    link
    fedilink
    English
    arrow-up
    1
    ·
    10 months ago

    Probably because every large LLM provider is trying to develop their own silicon to run their models.