langgenius / dify

Dify is an open-source LLM app development platform. Dify's intuitive interface combines AI workflow, RAG pipeline, agent capabilities, model management, observability features and more, letting you quickly go from prototype to production.
https://dify.ai
Other
52.36k stars 7.64k forks source link

Error about deploying locally a model plugged into the Hosted Inference API on huggingface #10984

Closed zhizhuaa closed 6 hours ago

zhizhuaa commented 7 hours ago

Self Checks

Dify version

0.10.2

Cloud or Self Hosted

Self Hosted (Docker)

Steps to reproduce

The following error occurred when I followed https://docs.dify.ai/zh-hans/development/models-integration/hugging-face to access the model from the community edition web page deployed via Docker Compose

image

Looking forward to your reply.

✔️ Expected Behavior

successfully access the model.

❌ Actual Behavior

The following error occurred when I followed https://docs.dify.ai/zh-hans/development/models-integration/hugging-face to access the model from the community edition web page deployed via Docker Compose

image

Looking forward to your reply.

zhizhuaa commented 7 hours ago

This is my first time using huggingface.

dosubot[bot] commented 7 hours ago

I couldn't find specific information about integrating Hugging Face models with Dify in the available data. However, there are some known issues and limitations with Dify version 0.10.2 when deploying models locally. Here are some general troubleshooting steps that might help:

  1. Check Network Configuration: Ensure that the Dify container can access the Hugging Face API. If you're using Docker, make sure the network settings allow for external API calls.

  2. Verify API Keys: Double-check that your Hugging Face API keys are correctly configured and have the necessary permissions.

  3. Inspect Docker Logs: Check the logs of your Docker containers for any error messages that might provide more insight into the issue.

  4. Cross-Domain Issues: If you encounter cross-domain issues, ensure that your domain configurations in docker-compose.yml are correct.

  5. Port Conflicts: Make sure there are no port conflicts that might be preventing the service from running properly.

If these steps don't resolve the issue, you might want to look into similar discussions or issues in the Dify repository for more context. For example, there are discussions about connection errors and deployment failures that might provide additional insights [1][2].

To continue talking to Dosu, mention @dosu.

crazywoola commented 6 hours ago

Please make sure you can access huggingface endpoint correctly.