-
Notifications
You must be signed in to change notification settings - Fork 149
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
[Bug] Better developer experience for bringing up TGI-Service #706
Comments
The service is still giving the same error after five hours:
Here are logs:
|
Hi @arun-gupta , the chatqna pipeline including tgi service can be started successfully on our xeon server. The root cause of your issue is The cache for model files in Transformers v4.22.0. 2024-08-30T19:00:17.058243Z WARN text_generation_launcher: No safetensors weights found for model Intel/neural-chat-7b-v3-3 at revision None. Converting PyTorch weights to safetensors. |
@letonghan my steps are available at https://gist.github.com/arun-gupta/7e9f080feff664fbab878b26d13d83d7. I can only use the published Docker images. What should I do differently? Tried with
|
This error is only occurring with Ubuntu 24.04 on AWS. I tested with both |
Hi @arun-gupta , since we don't have an AWS environment, currently it's hard to find out the root cause of this issue. |
@letonghan sure, let me set up a time with you offline. |
I tried this again on AWS Ubuntu 24.04 and it is working fine. It also worked with Ubuntu 24.04 on GCP with The bug can be closed. |
Ok, will close this issue. |
Priority
Undecided
OS type
Ubuntu
Hardware type
Xeon-SPR
Installation method
Deploy method
Running nodes
Single Node
What's the version?
0.9
Description
The instructions at https://github.com/opea-project/GenAIExamples/tree/main/ChatQnA/docker/xeon needs a better user experience.
Testing the LLM service says:
The container has been running for four hours now and still connecting to the service gives the following error:
There should be a clear indication of how the developer would know the download is finished. Also, the container name is
tgi-service
so that should be specified.Reproduce steps
The steps are documented at https://gist.github.com/arun-gupta/7e9f080feff664fbab878b26d13d83d7
Raw log
The text was updated successfully, but these errors were encountered: