open-compass / open-compass/VLMEvalKit

When running Qwen2.5-VL-72B on MUIRBench with 8 A100 GPUs, a CUDA out-of-memory error occurs.

Open
#1,110 2 comments 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

Dominant language
Python
Stars
4.4k
Forks
768
Avg merge
1d 10h
Merged PRs (30d)
17

Description

torch.OutOfMemoryError: CUDA out of memory. Tried to allocate 5.34 GiB. GPU 0 has a total capacity of 39.50 GiB of which 3.32 GiB is free. Including non-PyTorch memory, this process has 36.17 GiB memory in use. Of the allocated memory 35.53 GiB is allocated by PyTorch, and 141.78 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables)

How to Solve the CUDA Out-of-Memory Issue

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

Start by reproducing the Qwen2.5-VL-72B run on MUIRBench with 8 A100 GPUs and review the reported PyTorch CUDA memory-management guidance. Determine whether the run can complete without the out-of-memory error, documenting the configuration or change needed if it cannot.

Written by the indexing model from the issue text.

Assessment

Tech stack
python, pytorch
Domain
computer-vision, machine-learning
Issue type
Bug
Difficulty
4/5
Estimated time
3-5 days
Activity status
Stale
Clarity
Needs clarification
Newbie friendliness
25/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.