WARNING: THIS SITE IS A MIRROR OF GITHUB.COM / IT CANNOT LOGIN OR REGISTER ACCOUNTS / THE CONTENTS ARE PROVIDED AS-IS / THIS SITE ASSUMES NO RESPONSIBILITY FOR ANY DISPLAYED CONTENT OR LINKS / IF YOU FOUND SOMETHING MAY NOT GOOD FOR EVERYONE, CONTACT ADMIN AT ilovescratch@foxmail.com
Skip to content

[Bug]: 昇腾多机部署模型报Please don't use hard_fp4_kernels #114

@lanyanling111

Description

@lanyanling111

您的联系方式

[email protected]

问题描述

昇腾2机部署Qwen3-32B(每机1卡),命令为:
export WORLD_SIZE=2
export TASK_QUEUE_ENABLE=1
./script/ssh_multi_node.sh "host1,host2" 1 test/single_req_test.py models=Qwen3-32B models.ckpt_dir=/models/Qwen3-32B request.max_new_tokens=64 infer.cache_type=paged infer.tp_size=2 infer.attn_type=npu
报如下错误:
AssertionError: To chitu developers: Please don't use hard_fp4_kernels as an optional dependency name, it is not listed in get_requires.py.
[ERROR] 2025-11-15-16:24:10 (PID:2199, Device:-1, RankID:-1) ERR99999 UNKNOWN applicaiton exception
/usr/local/python3.11.13/lib/python3.11/tempfile.py:934: ResourceWarning: Implicitly cleaning up <TemporaryDirectory '/tmp/tmpsxnvo0ym'>
_warnings.warn(warn_message, ResourceWarning)

提交前请确认

  • 确认

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions