Get the latest tech news

g1: Using Llama-3.1 70B on Groq to create o1-like reasoning chains


g1: Using Llama-3.1 70b on Groq to create o1-like reasoning chains - bklieger-groq/g1

This is an experiment to show the power of prompting reasoning in visualized steps, not a comparison to or full replication of o1, which uses different techniques. In all-caps to improve prompt compliance by emphesizing the importance of the instruction, a set of tips and best practices are included. -> This helps the LLM remember to use techniques which produce better results, like breaking "strawberry" down into individual letters before counting.

Get the Android app

Or read this on Hacker News

Read more on:

Photo of Groq

Groq

Photo of Llama-3.1 70B

Llama-3.1 70B

Related news:

News photo

Valuation of AI chipmaking startup Groq soars to $2.8B after funding round

News photo

What AI bubble? Groq rakes in $640M to grow inference cloud

News photo

Groq secures $640M to supercharge AI inference with next-gen LPUs