minus-squarejun2san@alien.topBtoLocalLLaMA•NVidia H200 achieves nearly 12,000 tokens/sec on Llama2-13B with TensorRT-LLMlinkfedilinkEnglisharrow-up1·2 years agoHow much you want for your old H100? - me to ai devs linkfedilink
How much you want for your old H100? - me to ai devs