vxcontrol/pentagi
1
0
Fork 0
mirror of https://github.com/vxcontrol/pentagi.git synced 2026-07-20 12:37:04 -04:00
Code Issues 45 Packages Projects Releases 9 Wiki Activity

[GH-ISSUE #173] Problem: Conversation disconnection with the Baichuan Model. #69

New Issue
Closed
opened 2026-06-06 22:08:56 -04:00 by yindo · 3 comments
yindo commented 2026-06-06 22:08:56 -04:00
Owner
Copy Link
Copy Source

Originally created by @zh0604 on GitHub (Mar 3, 2026).
Original GitHub issue: https://github.com/vxcontrol/pentagi/issues/173

Image Image "During the process of adding a provider, connection tests sometimes fail with errors. Could this be a network problem? The earlier conversation interruption might be due to a provider-side connection disruption.
Originally created by @zh0604 on GitHub (Mar 3, 2026). Original GitHub issue: https://github.com/vxcontrol/pentagi/issues/173 <img width="1058" height="996" alt="Image" src="https://github.com/user-attachments/assets/913bf2be-0e52-4e48-bb2e-91267d7044f4" /> <img width="1125" height="1026" alt="Image" src="https://github.com/user-attachments/assets/e2dc8a45-49d8-40b2-99b0-cb18bc9d5ea7" /> "During the process of adding a provider, connection tests sometimes fail with errors. Could this be a network problem? The earlier conversation interruption might be due to a provider-side connection disruption.
yindo closed this issue 2026-06-06 22:08:56 -04:00
yindo commented 2026-06-06 22:08:57 -04:00
Author
Owner
Copy Link
Copy Source

@asdek commented on GitHub (Mar 4, 2026):

hey @zh0604

according to the screenshot you increased your quotes on the LLM provider but I think it's not a main reason, try to check your provider by console:
docker exec -it pentagi ./bin/ctester -agents all -verbose -groups all --workers 1

<!-- gh-comment-id:3995945080 --> @asdek commented on GitHub (Mar 4, 2026): hey @zh0604 according to the screenshot you increased your quotes on the LLM provider but I think it's not a main reason, try to check your provider by console: `docker exec -it pentagi ./bin/ctester -agents all -verbose -groups all --workers 1`
yindo commented 2026-06-06 22:08:57 -04:00
Author
Owner
Copy Link
Copy Source

@zh0604 commented on GitHub (Mar 5, 2026):

Image I have a .env file inside the pentagi folder, and I've already filled in the Baichuan model's key and URL into OPEN_AI_KEY and OPEN_AI_SERVER_URL. Why is it still throwing an error saying that the .env file is missing
<!-- gh-comment-id:4003448360 --> @zh0604 commented on GitHub (Mar 5, 2026): <img width="978" height="270" alt="Image" src="https://github.com/user-attachments/assets/c5eb6286-4b2d-4e11-9f85-8e862ba86958" /> I have a .env file inside the pentagi folder, and I've already filled in the Baichuan model's key and URL into OPEN_AI_KEY and OPEN_AI_SERVER_URL. Why is it still throwing an error saying that the .env file is missing
yindo commented 2026-06-06 22:08:57 -04:00
Author
Owner
Copy Link
Copy Source

@asdek commented on GitHub (Mar 19, 2026):

hey @zh0604

looking at your screenshots, i can see several configuration issues:

main problem: you're using OPEN_AI_KEY and OPEN_AI_SERVER_URL for Baichuan model, but these variables are only for the original OpenAI provider. for any other backend (like Baichuan), you must use LLM_SERVER_* variables instead.

rate limit errors (429): the errors in your second screenshot show "Request too frequent, please try again later" - this is OpenAI's rate limit, not a network issue. this confirms you're hitting OpenAI's API quota limits for your tier level.

configuration steps:

  1. clear OpenAI variables if you want to use Baichuan as primary provider:

    # remove or comment out:
# OPEN_AI_KEY=...
# OPEN_AI_SERVER_URL=...

  2. configure custom LLM provider for Baichuan:

    LLM_SERVER_URL=your_baichuan_api_url
LLM_SERVER_KEY=your_baichuan_api_key
LLM_SERVER_MODEL=your_model_name

  3. configure embedding provider (required, you can keep OpenAI for embeddings with tier 1):

    OPEN_AI_KEY=your_openai_key  # keep this for embeddings only
EMBEDDING_PROVIDER=openai
EMBEDDING_MODEL=text-embedding-3-small # or text-embedding-3-large

see the embedding configuration guide for details.

about tier limits: according to your screenshots, the 429 errors indicate you exceeded your OpenAI rate limits. the product will work with tier 1, but very slowly due to rate limiting. we recommend:

  • tier 3+ for comfortable OpenAI usage
  • alternative providers like Moonshot, Alibaba Cloud (Qwen), z.ai (GLM), DeepSeek - these have different pricing and quotas
  • hybrid setup: use tier 1 OpenAI for embeddings only + another provider for main LLM inference

all provider configurations are tested in the latest build, so i'm closing this issue for now. if you have updates or can't configure it correctly after following these steps, feel free to reopen with new information!

<!-- gh-comment-id:4094142225 --> @asdek commented on GitHub (Mar 19, 2026): hey @zh0604 looking at your screenshots, i can see several configuration issues: **main problem**: you're using `OPEN_AI_KEY` and `OPEN_AI_SERVER_URL` for Baichuan model, but these variables are **only for the original OpenAI provider**. for any other backend (like Baichuan), you must use `LLM_SERVER_*` variables instead. **rate limit errors (429)**: the errors in your second screenshot show "Request too frequent, please try again later" - this is **OpenAI's rate limit**, not a network issue. this confirms you're hitting OpenAI's API quota limits for your tier level. **configuration steps**: 1. **clear OpenAI variables** if you want to use Baichuan as primary provider: ```bash # remove or comment out: # OPEN_AI_KEY=... # OPEN_AI_SERVER_URL=... ``` 2. **configure custom LLM provider** for Baichuan: ```bash LLM_SERVER_URL=your_baichuan_api_url LLM_SERVER_KEY=your_baichuan_api_key LLM_SERVER_MODEL=your_model_name ``` 3. **configure embedding provider** (required, you can keep OpenAI for embeddings with tier 1): ```bash OPEN_AI_KEY=your_openai_key # keep this for embeddings only EMBEDDING_PROVIDER=openai EMBEDDING_MODEL=text-embedding-3-small # or text-embedding-3-large ``` see the [embedding configuration guide](https://github.com/vxcontrol/pentagi?tab=readme-ov-file#embedding-configuration-and-testing) for details. **about tier limits**: according to your screenshots, the 429 errors indicate you exceeded your OpenAI rate limits. the product will work with tier 1, but **very slowly** due to rate limiting. we recommend: - **tier 3+** for comfortable OpenAI usage - **alternative providers** like Moonshot, Alibaba Cloud (Qwen), z.ai (GLM), DeepSeek - these have different pricing and quotas - **hybrid setup**: use tier 1 OpenAI for embeddings only + another provider for main LLM inference all provider configurations are tested in the latest build, so i'm closing this issue for now. if you have updates or can't configure it correctly after following these steps, feel free to reopen with new information!
Sign in to join this conversation.
No Branch/Tag Specified
Branches Tags
Labels
Clear labels
bug
enhancement
help wanted
pull-request
Mirrored from GitHub Pull Request
 question
No Label
Milestone
No items
No Milestone
Projects
Clear projects
No project
Assignees
Clear assignees
yindo
No Assignees
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: vxcontrol/pentagi#69
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?