lm-sys / lm-sys/FastChat

Out of physical memory when loading model

Open
#521 0 comments 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

Dominant language
Python
Stars
39.5k
Forks
4.8k
PR merge metrics
No merged PRs in 30d

Description

Hi, I got the following error when loading llama-7B model (ported to huggingface), my server has 256GB RAM on board. Is there any option to reduce the RAM consumption? Thanks.
`fatal : Memory allocation failure`
or
`ERROR:torch.distributed.elastic.multiprocessing.api:failed (exitcode: -9) local_rank: 5 (pid: 32708) of binary: /home/user/miniconda3/envs/pytorch1.12/bin/python`

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.

Research direction

The report names llama-7B loading, a Hugging Face port, and PyTorch distributed failures, but no repository file, command, or test. Start by collecting the model-loading and distributed-launch steps and measuring peak RAM; done would require a reproducible diagnosis and a documented or verified way to reduce memory consumption.

Written by the indexing model from the issue text.

Assessment

Tech stack
huggingface, python, pytorch
Domain
machine-learning
Issue type
Bug
Difficulty
5/5
Estimated time
Over a week
Activity status
Stale
Clarity
Needs clarification
Newbie friendliness
20/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.