microsoft / microsoft/MInference

[Question]: Discrepancy in Pre-filling Time and Memory Consumption on Single A100

Open
#84 3 comments 0 reactions 1 assignee View on GitHub

@iofu728 is already working on this.

Since Nov 18, 2024.

question
Dominant language
Python
Stars
1.2k
Forks
82
Avg merge
1d 18h
Merged PRs (30d)
1

Description

Describe the issue

I came across your statement in the paper where you mentioned:

"When serving LLaMA-3-8B on a single A100 machine, the model would keep users waiting for 6 minutes to finish the pre-filling stage given a prompt of 300K tokens, and this number increases to 30 minutes for a prompt of 1M tokens."

However, I am also running on a single A100 (80GB) and using Hugging Face's implementation of LLaMA in SDPA mode. With a 50k token context, the pre-fill time is around 2.5 seconds, but when using 100k tokens, I run into an "Out of Memory" issue.

Could you clarify why there is such a significant discrepancy between your results and mine? Is there something I might be missing or misunderstanding?

Thanks for your help!

Contributor guide

No contributing guide indexed for this repository

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

Assessment

This issue has not been assessed yet.

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.