Content
@
0 reply
0 recast
0 reaction
gm8xx8
@gm8xx8
apparently Llama 3 now at up to 350-380 tokens per second for Llama 3 8B and up to 150 tokens per second for Llama 3 70B. qroq ish ✔️
4 replies
1 recast
6 reactions
gm8xx8
@gm8xx8
having fun in the playground. https://www.together.ai/blog/together-ai-partners-with-meta-to-release-meta-llama-3-for-inference-and-fine-tuning in the middle of digging deep into this 😈
0 reply
0 recast
0 reaction
@king
man zuck really went super saiyan on this
0 reply
0 recast
1 reaction
kevin j 🤗
@entropybender
do you know on which hardware this was benchmarked on
0 reply
0 recast
0 reaction
