What is Azure thinking with this VM size NVadsA10 v5-serieslearn.microsoft.com1 points·by ssheng·2 jaar geleden·1 comments
ssheng·2 jaar geleden·discussTo get a full A10, one must also get 36 vCPU and 440 GB of memory. I must be missing something.
ssheng·2 jaar geleden·discussQuality loss with quantization is expected. It seems like with GPTQ the loss is within acceptable range based on the perplexity score shown.
ssheng·2 jaar geleden·discussCreative idea. Any data on how much it takes to load the LoRAs and how much latency it adds to the generation speed?
ssheng·2 jaar geleden·discussHello! We're the authors of this blog post. Please let us know if there are other models and inference backends you'd like us to benchmark next.