rihard7854@alien.topB to LocalLLaMAEnglish · 2 years agoNVidia H200 achieves nearly 12,000 tokens/sec on Llama2-13B with TensorRT-LLMgithub.comexternal-linkmessage-square23linkfedilinkarrow-up11arrow-down10
arrow-up11arrow-down1external-linkNVidia H200 achieves nearly 12,000 tokens/sec on Llama2-13B with TensorRT-LLMgithub.comrihard7854@alien.topB to LocalLLaMAEnglish · 2 years agomessage-square23linkfedilink
minus-squareLongjumping-Bake-557@alien.topBlinkfedilinkEnglisharrow-up1·2 years agoAnd that’s on a die just slightly bigger than the 4090. Unless they increased the size compared to h100?
And that’s on a die just slightly bigger than the 4090. Unless they increased the size compared to h100?