Ruonan Wang

> When “ollama run gemma3n:e4b“ in [ollama-ipex-llm-2.3.0b20250630]， I got： Error: llama runner process has terminated: exit status 2 Hi @yizhangliu , could you please provide us with detailed server log...

Hi @savvadesogle , I didn't reproduce the error on my Linux machine, on my side, `gemma3n:e4b` / `gemma3n:e2b` both works with latest ollama portable zip, and based on my observation,...

@savvadesogle @yizhangliu It seems you are facing the same error. Have you ever run other ollama models successfully or run other version ollama portable zip successfully on your machine ?...

Hi @FilipLaurentiu , we never meet such error before. What is your CPU & GPU ? And have you ever run other ollama models successfully or run other version ollama...

@savvadesogle @yizhangliu we have made a fix for gemma3n, you could try if this new release (https://github.com/ipex-llm/ipex-llm/releases/download/v2.3.0-nightly/ollama-ipex-llm-2.3.0b20250710-ubuntu.tgz / https://github.com/ipex-llm/ipex-llm/releases/download/v2.3.0-nightly/ollama-ipex-llm-2.3.0b20250708-win.zip) works.

> If I don't specify the export SYCL_CACHE_PERSISTENT=1 variable, then ollama works. Yes, `unset SYCL_CACHE_PERSISTENT` can solve this issue, we will consider to update our quickstart & script later. 😊...

Hi @hurui200320 , with https://github.com/ipex-llm/ipex-llm/releases/download/v2.3.0-nightly/ollama-ipex-llm-2.3.0b20250630-ubuntu.tgz, `ollama run deepseek-r1:8b` works fine on linux ARC dGPU. ```bash arda@arda-arc:~/ollama-ipex-llm-2.3.0b20250630-ubuntu$ ./ollama run deepseek-r1:8b >>> hi Thinking... 嗯，用户只发了一个简单的问候“hi”，看起来像是想开启对话但还没想好具体要说什么。这种打招呼的方式很常见，可能ta刚打开聊天界面还在试探功能边 界，或者单纯想测试响应速度。 考虑到这是典型的非正式开场白，需要保持友好轻松的氛围来降低用户的紧张感。用表情符号可以快速建立亲和力，而开放式提问能给用户足够的接话空间 ——既不能预设太多限制让ta觉得被绑架，又要比机械回复多提供几个认知锚点。 选项设计成生活化场景比较好：学习、工作、闲聊分别覆盖了求知型和技术型用户的潜在需求。用“随便聊聊”对应情感需求，避免把所有对话都框定在解决 问题的模式里。最后加个爱心符号强化温暖感——毕竟冰冷的知识库加上笑脸总比单纯问答更有人情味。 用户此刻可能处于两种状态：要么是带着明确目标来闲聊（比如想测试AI反应），要么真的没头绪只是随手发了个消息。所以回复要像打开门一样，既欢迎 有目的来的访客，也接纳迷路的路人。...

Hi @LIwii1 I think this error is caused by out of GPU memory. Deepseek-r1:8b-0528-qwen3-fp16 model itself requires 16GB. You could try other precision instead.

Hi @doublefx , could you please try `conda install libuv` to see if it helps ? Just in case you miss this step during create your conda env.

Hi @doublefx , I think I need to make some clarification. `ipex-llm[xpu]` and `ipex-llm[npu]` are different packages which target different hardwares. If you want to run on GPU, you should...