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

Y	Hacker News new \| ask \| show \| jobs

	Show HN: Beating OpenAI's structured outputs on cost, accuracy and speed (boundaryml.com)
	8 points by hellovai 669 days ago

1 comments

ddematheu 669 days ago

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

link

hellovai 669 days ago

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!

link