Predibase (@predibase) 's Twitter Profile
Predibase

@predibase

GPT-4 Quality for Less than GPT-3.5 Prices

The fastest, most efficient way to fine-tune and serve small models for your use case—without sacrificing quality.

ID: 1333528002907361281

linkhttps://predibase.com/ calendar_today30-11-2020 21:47:24

339 Tweet

1,1K Followers

9 Following

Predibase (@predibase) 's Twitter Profile Photo
Predibase
@predibase

2 months ago

There's a new "best #SLM" in town! We fine-tuned Llama-3.1-8b-instruct on 25 tasks and it shows a huge improvement over #GPT-4, GPT-4o mini, fine-tuned #Phi-3, and fine-tuned #Mistral-7b. Small language models continue to set the standard for performance, cost, and privacy!

There's a new "best #SLM" in town! We fine-tuned Llama-3.1-8b-instruct on 25 tasks and it shows a huge improvement over #GPT-4, GPT-4o mini, fine-tuned #Phi-3, and fine-tuned #Mistral-7b. Small language models continue to set the standard for performance, cost, and privacy!