Meta’s benchmarks for its new AI models are a bit misleading

  • staffstaff
  • AI
  • April 6, 2025
  • 0 Comments

One of the new flagship AI models Meta released on Saturday, Maverick, ranks second on LM Arena, a test that has human raters compare the outputs of models and choose which they prefer. But it seems the version of Maverick that Meta deployed to LM Arena differs from the version that’s widely available to developers. […]

  • Related Posts

    OpenAI co-founder Ilya Sutskever’s Safe Superintelligence reportedly valued at $32B

    Safe Superintelligence (SSI), the AI startup led by OpenAI’s co-founder and former chief scientist Ilya Sutskever, has raised an additional $2 billion in funding at a $32 billion valuation, according…

    Continue reading
    The xAI–X merger is a good deal — if you’re betting on Musk’s empire

    When Elon Musk announced that his AI startup, xAI, had acquired his social media company, X (formerly known as Twitter), in an all-stock deal, it raised some eyebrows. But in…

    Continue reading

    Leave a Reply

    Your email address will not be published. Required fields are marked *

    You Missed

    OpenAI co-founder Ilya Sutskever’s Safe Superintelligence reportedly valued at $32B

    • By staff
    • April 12, 2025
    • 2 views

    Apple’s ‘Mythic Quest’ is ending with an updated Season 4 finale

    • By staff
    • April 12, 2025
    • 2 views

    The xAI–X merger is a good deal — if you’re betting on Musk’s empire

    • By staff
    • April 12, 2025
    • 3 views

    Meta’s vanilla Maverick AI model ranks below rivals on a popular chat benchmark

    • By staff
    • April 11, 2025
    • 4 views