Skip to content

[Bug]: The Qwen3 - Next - 80B - A3B - Instruct model is deployed on 8 cards of the 910B with 64G memory, and out - of - memory (OOM) occurs. #4379

@KannibalH

Description

@KannibalH

Your current environment

vllm-ascend

version:v0.11.0rc2-openeuler

Docker Compose

Image

🐛 Describe the bug

Docker Container Logs

Image Image Image

Metadata

Metadata

Assignees

No one assigned

    Labels

    bugSomething isn't working

    Type

    No type

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions