Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Failed to load model from file: /data/DeepSeek R1-Distill-Qwen-32B-Q4 K M.gguf #2830

Open
1 of 3 tasks
thymol-zzw opened this issue Feb 10, 2025 · 6 comments
Open
1 of 3 tasks
Milestone

Comments

@thymol-zzw
Copy link

System Info / 系統信息

xinference v1.2.1
Failed to load model from file: /

data/DeepSeek R1-Distill-Qwen-32B-Q4 K M.gguf

Running Xinference with Docker? / 是否使用 Docker 运行 Xinfernece?

  • docker / docker
  • pip install / 通过 pip install 安装
  • installation from source / 从源码安装

Version info / 版本信息

xinference v1.2.1

The command used to start Xinference / 用以启动 xinference 的命令

xinference v1.2.1

Reproduction / 复现过程

Image

Expected behavior / 期待表现

成功运行DeepSeek R1-Distill-Qwen-32B-Q4 K M.gguf

@XprobeBot XprobeBot added this to the v1.x milestone Feb 10, 2025
@qinxuye
Copy link
Contributor

qinxuye commented Feb 10, 2025

llama-cpp-python 升级到 0.3.7 试下。

@frankjoey2048
Copy link

llama-cpp-python 升级到 0.3.7 试下。
这个是docker的问题,应该重新打包。

@tigflanker
Copy link

尝试容器内升级了一下llama-cpp-python,gguf可以成功加载了,但是因为我升级的方式不对,gguf是按照CPU加载的

  1. 登录容器后:
    root@b47fbfcc428b:/opt/inference# pip config set global.index-url https://pypi.mirrors.ustc.edu.cn/simple
    root@b47fbfcc428b:/opt/inference# pip install --upgrade llama-cpp-python

我尝试过:
root@b47fbfcc428b:/opt/inference# CMAKE_ARGS="-DGGML_BLAS=ON -DGGML_BLAS_VENDOR=OpenBLAS" pip install --upgrade llama-cpp-python
但是cpp编译失败,真的llama-cpp-python每次编译都不成功

  1. exit退出容器后:
    对容器commit,然后按照这个新镜像启动
    docker commit <container_id> <new_image_name>:

@jacktpy
Copy link

jacktpy commented Feb 12, 2025

尝试容器内升级了一下llama-cpp-python,gguf可以成功加载了,但是因为我升级的方式不对,gguf是按照CPU加载的

  1. 登录容器后:
    root@b47fbfcc428b:/opt/inference# pip config set global.index-url https://pypi.mirrors.ustc.edu.cn/simple
    root@b47fbfcc428b:/opt/inference# pip install --upgrade llama-cpp-python

我尝试过: root@b47fbfcc428b:/opt/inference# CMAKE_ARGS="-DGGML_BLAS=ON -DGGML_BLAS_VENDOR=OpenBLAS" pip install --upgrade llama-cpp-python 但是cpp编译失败,真的llama-cpp-python每次编译都不成功

  1. exit退出容器后:
    对容器commit,然后按照这个新镜像启动
    docker commit <container_id> <new_image_name>:

我也是一样的,一直编译失败

@tigflanker
Copy link

尝试容器内升级了一下llama-cpp-python,gguf可以成功加载了,但是因为我升级的方式不对,gguf是按照CPU加载的

  1. 登录容器后:
    root@b47fbfcc428b:/opt/inference# pip config set global.index-url https://pypi.mirrors.ustc.edu.cn/simple
    root@b47fbfcc428b:/opt/inference# pip install --upgrade llama-cpp-python

我尝试过: root@b47fbfcc428b:/opt/inference# CMAKE_ARGS="-DGGML_BLAS=ON -DGGML_BLAS_VENDOR=OpenBLAS" pip install --upgrade llama-cpp-python 但是cpp编译失败,真的llama-cpp-python每次编译都不成功

  1. exit退出容器后:
    对容器commit,然后按照这个新镜像启动
    docker commit <container_id> <new_image_name>:

我也是一样的,一直编译失败

这个llama_cpp_python,我感觉是我安装包里面难安装的,可能还是没摸清门道,按照官网的介绍每次编译不成功

刚刚从dockerfile里面翻了一下:
root@d4d540eaf808:/opt/inference# pip install "llama-cpp-python>=0.3.7" -i https://abetlen.github.io/llama-cpp-python/whl/cu124Looking in indexes: https://abetlen.github.io/llama-cpp-python/whl/cu124

也还没有造好的轮子,可能得git源码包进行编译
ERROR: Could not find a version that satisfies the requirement llama-cpp-python=0.3.7 (from versions: 0.2.66, 0.2.67, 0.2.68, 0.2.69, 0.2.70, 0.2.71, 0.2.72, 0.2.73, 0.2.74, 0.2.75, 0.2.76, 0.2.77, 0.2.81, 0.2.82, 0.2.83, 0.2.84, 0.2.85, 0.2.86, 0.2.87, 0.2.88, 0.2.90, 0.3.4)
ERROR: No matching distribution found for llama-cpp-python=0.3.7

@qinxuye
Copy link
Contributor

qinxuye commented Feb 12, 2025

llama-cpp-python 跟不上 llama.cpp 的发展了。我们计划自己来做 llama.cpp 的封装。cc @codingl2k1

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

6 participants