CodeWithLLM-Updates
-

🥳 Llama 3 is out!

Models were trained on two newly announced clusters with specially built 24K GPUs on over 15T tokens of data - the training dataset is 7 times larger than that used for Llama 2,
including 4 times more code.

https://llama.meta.com/llama3/

models are also deployed on LPU https://groq.com/