Groq enables much faster throughput at the lowest cost possible! Almost 500 tokens per second for lower cost!

but what’s the quality? it’s not on any of the quality lists

🇧🇷🇺🇸-  Book: Making Things Think: https://holloway.com/mtt. Investor in Wander, Carry, Footprint, Merkle Manufactory (Farcaster), Dynamic, Paragraph.

/bettik founder /outercorps stalwart | ex-binance, ex-celsiusX

i didn't know that. this comparison lists them separately with unknown # of parameters for grok, but it knows that value for Mixtral 8x7B?

i didn't know that. this comparison lists them separately with unknown # of parameters for grok, but it knows that value for Mixtral 8x7B?
https://zapier.com/blog/best-llm/

Grok and Groq are different things.
Grok is LLM by Twitter/X. 
Groq mentioned above is startup that has custom GPUs and lets you run different open source models, like Mixtral 8x7B or Llama2. 

Point is that using Groq API you can have very fast inference with these models vs using competitors like perplexity.