mirror of
https://github.com/vxcontrol/pentagi.git
synced 2026-07-21 08:15:23 -04:00
[GH-ISSUE #27] [Bug]: missing deepseek.provider.yml #12
Reference in New Issue
Block a user
Delete Branch "%!s()"
Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?
Originally created by @dominyko on GitHub (Mar 31, 2025).
Original GitHub issue: https://github.com/vxcontrol/pentagi/issues/27
Originally assigned to: @asdek on GitHub.
Affected Component
External Integrations (LLM/Search APIs)
Describe the bug
Hello there,
I'm having issues with using deepseek API, while it works with OpenAI.
and tailing pentagi container shows me this error:
failed to initialize providers: failed to create custom provider: failed to load provider config: failed to read config file: open /opt/pentagi/conf/deepseek.provider.yml: no such file or directory.
Also "custom" is selected at the top left.
Steps to Reproduce
Custom LLM provider
LLM_SERVER_URL=https://api.deepseek.com
LLM_SERVER_KEY=sk-or-vxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx
LLM_SERVER_MODEL=
LLM_SERVER_CONFIG_PATH=/opt/pentagi/conf/deepseek.provider.yml
I tried to insert my model in LLM_SERVER_MODEL, but it did not have any effect.
System Configuration
Latest version PentAgi
Docker version 28.0.4, build b8034c0
kali-linux-2025.1a-vmware-amd64
4GB RAM
4 Cores
80 GB
Logs and Artifacts
└─$ docker logs pentagi
service ssl crt and key already exist
2025/03/31 16:56:33 goose: no migrations to run. current version: 20250322
2025/03/31 16:56:33 Migrations ran successfully
2025/03/31 16:56:33 failed to initialize providers: failed to create custom provider: failed to load provider config: failed to read config file: open /opt/pentagi/conf/deepseek.provider.yml: no such file or directory
service ssl crt and key already exist
Screenshots or Recordings
I confirm that the file does not exist.

Verification
@asdek commented on GitHub (Mar 31, 2025):
Hello @dominyko
Thank you for reporting this issue, it helps us improve the product! 😉
You are absolutely correct. There are missed adding the line in the Dockerfile to copy the
deepseek.provider.ymlfile into the final Docker image. I've fixed this in the commit: ced7b1a.To resolve the problem, please update your Docker image and restart the container by running:
P.S. sorry for so stupid bug... 😔
@summersblue commented on GitHub (Apr 1, 2025):
I met the same problem, I resolved it by change docker-compose.yaml volumes:
- pentagi-data:/opt/pentagi/data
- pentagi-ssl:/opt/pentagi/ssl
- /var/run/docker.sock:/var/run/docker.sock
- /mnt/pentagi/conf/deepseek.provider.yml:/opt/pentagi/conf/deepseek.provider.yml # add this line,others keep the original
@asdek commented on GitHub (Apr 2, 2025):
Hello @summersblue
Yes, that's a valid solution, and it's correct if you want to customize the
deepseek.provider.ymlfile. You can mount it into the same directory with any name and specify its path using theLLM_SERVER_CONFIG_PATHenvironment variable. However, if you modify this file, I strongly recommend running the testing utility to ensure your LLM settings pass all the tests. This will help guarantee the agents operate reliably.You can run the testing utility with the following command:
For more details, please refer to the Running Tests in a Production Environment section in the README.
Alternatively, in commit ced7b1a, the original
deepseek.provider.ymlfile from the repository was added to the Docker image. This means you can simply update your Docker image and restart the container without needing to manually mount the configuration file:Thank you for your feedback! Your contributions help make the product better for everyone. 😉