✅WizardCoder-34B surpasses GPT-4, ChatGPT-3.5 and Claude-2 on HumanEval with 73.2% pass@1

  • Anony Moose
    link
    fedilink
    English
    arrow-up
    1
    ·
    1 year ago

    Agreed, but still huge progress in OSS models in a very short time!