Skip to content

Release v1.1
Compare
Choose a tag to compare
@nlpxucan nlpxucan released this 26 May 15:53
· 371 commits to main since this release
v1.1
0966834
This commit was created on GitHub.com and signed with GitHub’s verified signature. The key has expired.
GPG key ID: 4AEE18F83AFDEB23
Expired
Learn about vigilant mode.

🚀Major Update: Introducing WizardLM 13B Version.

  • On difficulty-balanced Evol-Instruct testset, evaluated by GPT-4: WizardLM-13B achieves 89.1% of ChatGPT, Vicuna-13B achieves 86.9%, and WizardLM-7B achieves 78%.
  • The 13B version is trained on instruction data evolved from real-world human conversations (ShareGPT), while the 7B version is trained on instruction data evolved from machine-generated data (Alpaca).
  • We provide a comparison between the performance of the WizardLM-13B and ChatGPT on different skills to establish a reasonable expectation of WizardLM's capabilities.
Assets 2
Loading