Show HN: Beating OpenAI's structured outputs on cost, accuracy and speed

Posted by hellovai on Wednesday August 14, 2024 @03:02PM
from the show-hn dept.

https://www.boundaryml.com/blog/sota-function-calling

( 2 comments | 3 points | on HN )

Re:Show HN: Beating OpenAI's structured outputs on cost, acc... by ddematheu on Wednesday August 14, 2024 @05:12PM (#41250553)

Interesting performance with GPT 3.5, what does performance on Llama look like? What about smaller models like Llama 3.1?

[ Parent | on HN ]

Re:Show HN: Beating OpenAI's structured outputs on cost, acc... by hellovai on Wednesday August 14, 2024 @05:40PM (#41250786)

We have some preliminary data with llama3.1 and we find that the smaller model gets to around 70% with BAML (+20% from base), but we'll update this dashboard with llama3.1 by end of week!

[ Parent | on HN ]