Artificial Intelligence (AI)

Llama 3 70B on Groq is:

~17x lower input cost 🤯

~38x lower output cost 🤯🤯

14x faster @ ~280 vs ~20 tokens per sec 🤯🤯🤯

Compared to OpenAI GPT 4

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

Haven’t dug much, do they have special hardware or just powerful inference engine?

It’ll be onchain soon, giving smart contracts the power of 70B parameters of computation & intelligence