Welcome to AI-Playground Discussions!

[Nuullll]

👋 Welcome!

We’re using Discussions as a place to connect with other members of our community. We hope that you:

• Ask questions you’re wondering about.
• Share ideas.
• Engage with other community members.
• Welcome others and are open-minded. Remember that this is a community we
  build together 💪.

To get started, comment below with an introduction of yourself and tell us about what you do with this community.

[bobduffy]

Thanks for adding. Adding a link here to our Discord channels on AI Playground

[brownplayer]

An error occurred while loading mistral-nemo. What is the cause

[majerus1223]

hey

[lh3953]

在打开程序的时候 一直提示AI playground loading，怎么解决？

[brownplayer]

按照官方的说法就是

这个卸了重新装

[Nuullll]

请根据这里的步骤查看完整错误信息，大概率是因为网络问题环境没安装好。

[yeetecer]

Backend initialization failed
The AI inference backend failed to initialize. Please try restarting the application. If the problem persists, you can check the Details for additional information about the error.

[lh3953]

我已经尝试重新安装这个软件，也重新安装过系统，还重新安装了显卡驱动。但是还是没办法解决，这个问题的根源是软件安装失败？还是我电脑网络问题？或者是我的系统（win10）不能使用这个app？

[shira-zusa]

which serials GPU do you use? I can run it with B580 and amd 5600

[FAIOJS]

组件安装特别慢，有什么方法可以加快吗？

[Xtiger909]

如何修改模型的默认目录？ 为何自行指定模型目录后就提示错误，路径只有英文。
How to modify the default directory of the model? Why does specifying a custom model directory result in an error, and why does the path only accept English characters?

[WeihanLi]

Would be great to support Ollama compatible API and the winget package install/upgrade

[WeihanLi]

seeing that there's a portable ollama, not sure if it could be integrated with the AI-Playground

https://github.com/intel/ipex-llm/blob/main/docs/mddocs/Quickstart/ollama_portable_zip_quickstart.md

[zoner72]

it is so fast! even on my A380
Is there a server API, eg rest interface localhost:port?

[shira-zusa]

I want change my image save path,but i dont how to do?someone can heple me,thanks.

[lh3953]

cpu：13600KF, mobo：B660..  GPU:A770     The GPU is also produced by Intel itself, which is why I want to use this software. However, I always can't make it work.After installing the latest version, I have now resolved the issue of being unable to log in. The new problem is that I can't download the large model. In the software interface, there is a prompt asking me to download the model file, but the size of the model doesn't show up, and I can't click to download it. I thought it was a problem with my network, but even after using a VPN, it still doesn't work.

[bobduffy]

I'll alert the dev team. I believe there should be a time out that allows you to confirm the terms under a slow network route

[lh3953]

应该是能用，我试一下。主要是这个主程序无法在程序内直接下载模型，点击下载的按钮直接就是灰色的，我用了vpn和加速器都不行。比较烦。。

[shira-zusa]

what happening with my GPU? could someone help me fix it

[salmonlips]

I'm loving your software, i just got gpt-oss-20b up running, is ther a way to hide the thinking context or minimize it, and then have the output? or is that in the plans?
it's a bit difficult parsing where it stops thinking and where it starts replying

[salmonlips]

side note on that same subject within the gguf tab i can't load documents to refernece either they just spin out stall, openvino is fine but basically can only get llama 3.1 running of anything newer

[bobduffy]

When models reason, you can hide the reasoning. GPT-OSS sometimes carries reasoning to its answer.

For embedding document first try something simple. Put your name in a .txt file. Attach and see ask it what your name is.

Then expand from there

[salmonlips]

that helped, i find gpt is always carrying reasoninginto the answer and then putting out the answer, it's a good answer much better than llama and phi haha but it's, still hard to parse
you ended up being right about the txt file as well i think i was throwing something too big at it, in the openvino side of the models it could handle it no issue but on this side its' nto happening

[gnaw0725]

When I try to install the AI Playground component, it consistently fails with the following error message. My graphics card is an Intel Arc, and both the GPU and NPU drivers have been updated to the latest version.

[device-service] {
"device_list": [
{
"device_id": 0,
"device_name": "Intel(R) Arc(TM) Graphics",
"device_type": "GPU",
"pci_bdf_address": "0000:00:02.0",
"pci_device_id": "0x7d55",
"uuid": "00000000-0000-0200-0000-00087d558086",
"vendor_name": "Intel(R) Corporation"
}
]
}

index-wNbU7XkZ.js:156 [ai-backend] Couldn't detect best device, selecting first. Error: TypeError: Cannot read properties of undefined (reading 'id')
（匿名） @ index-wNbU7XkZ.js:156
（匿名） @ VM5:2
emit @ VM4 sandbox_bundle:2
onMessage @ VM4 sandbox_bundle:2
index-wNbU7XkZ.js:13 Error: Error invoking remote method 'detectDevices': TypeError: Cannot read properties of undefined (reading 'name')
zR @ index-wNbU7XkZ.js:13
Xo @ index-wNbU7XkZ.js:13
（匿名） @ index-wNbU7XkZ.js:13
Promise.catch
cr @ index-wNbU7XkZ.js:13
cr @ index-wNbU7XkZ.js:13
n @ index-wNbU7XkZ.js:17

[bobduffy]

Submit your issue to the issues tab. Provide information about your system. Also the true issue might be an error just before this output

[calvinwonghk]

does it support intel core ultra 9 285k?