Tool Calling Showdown: Llama3.1 8B vs 70B - Shocking Results Revealed!

Описание к видео Tool Calling Showdown: Llama3.1 8B vs 70B - Shocking Results Revealed!

In this video, I dive deep into the function calling capabilities of Llama 3.1 models, comparing the 8 billion and 70 billion parameter versions. Using my custom test suite, I generate 100 different sentences to rigorously test each model. The results are shocking: the 8B model struggles significantly, while the 70B model excels with over 95% accuracy! Join me as I walk through the Python code used for these tests and uncover the surprising differences in performance. Perfect for AI enthusiasts and developers looking to understand model capabilities!

🔗 Links & Resources:
Github:
https://github.com/Teachings/langgrap...

Watch full series on LangGraph here:
   • AI Workflows: LangGraph and RAG Series  

Комментарии

Информация по комментариям в разработке

Похожие видео