[samosx]

Having a hard time with estimating how much GPU memory that LLM needs to serve it? What kind of GPUs to use and how many?

Wrote a blog post to demystify the process of GPU memory usage estimating.

[brianjking]

My issue is figuring out how to identify how many concurrent users you can support on average on a given GPU.

Understanding the vram to simply load the weights is easy enough. When you are allowing for something like content generation with varying lengths of input/output tokens, how do you even begin to identify the GPUs you need?