dayskk

Results 4 issues of dayskk

ipex-llm报错如图片server cpu not listed in available servers map

8 comment

![微信图片_20240627142704](https://github.com/intel-analytics/ipex-llm/assets/166265863/13764abc-6586-43cc-8ecf-08bb042f194c) ![微信图片_20240627142723](https://github.com/intel-analytics/ipex-llm/assets/166265863/c88be310-41c4-40b0-bff6-f0f5cdcd47e2) ![微信图片_20240627142727](https://github.com/intel-analytics/ipex-llm/assets/166265863/ebc91b91-d058-4c72-9451-bf90ebafeab3) 用ollama qwen2:7b

user issue

intel gpu and ollama error

3 comment

![微信图片_20240605135354](https://github.com/intel-analytics/ipex-llm/assets/166265863/4bcfc12a-ead8-468a-ab24-dfe60fb1d9d4) The following error occurred after running for a period of time, please refer to the attachment. Currently, no reproduction method has been found. GPU accelerates Ollama to run Qwen1.5...

user issue

要节约磁盘空间，ollama 使用igpu时能不能把oneapi必要库拷贝到conda环境

4 comment

要节约磁盘空间，ollama 使用igpu时能不能把oneapi必要库拷贝到conda环境，以下是收到的邮件方法，但是不适用于ollama，没有intel-extension-for-pytorch文件夹，torch的lib现成就有文件了，都不用拷贝 Hi Xutao, 在你们参展的MTL Ultra7 机器可以把oneapi的必要库拷贝到conda，节约机器磁盘空间，也不需要每次启动llm应用前运行 oneapi setvars.bat. 步骤： （1）从C:\Program Files (x86)\Intel\oneAPI把以下这些dll拷贝到conda环境的intel-extension-for-pytorch/bin 路径下 （2）从C:\Program Files (x86)\Intel\oneAPI把以下这些dll拷贝到conda环境的torch/lib 路径下

user issue

llama runner process has terminated exit status 1

6 comment

用ollama run qwen2:7b 还是用curl模式，都返回错误llama runner process has terminated exit status 1 ![微信图片_20240716184524](https://github.com/user-attachments/assets/ba6ee0cb-6e30-4cfd-ba5c-0c4fd5a7446a) ![微信图片_20240716184552](https://github.com/user-attachments/assets/b683dd10-6b39-4a8d-8284-e3eeb37b2aa4) ![微信图片_20240716184749](https://github.com/user-attachments/assets/b0c9899b-ab27-4f68-90c1-62bfbf6fa53e)

user issue