Error with ollama and openai adapters

[matiasmolinas]

Hi, after trying the latest version of bee-stack with Ollama and OpenAI, I am unable to use Bee-Stack on an Ubuntu machine with 32GB of RAM. I encountered different errors in each case. I have verified that Ollama is running properly and that the OpenAI key is functioning correctly.

Attached are screenshots and logs for each case.


2025-01-16_1004S.zip

[Tomas2D]

Please try to pull the latest `bee-stack and run the following commands

./bee-stack.sh clean
./bee-stack.sh setup
./bee-stack.sh start

[Tomas2D]

The default model that API uses is llama3.1, ensure you have it; run ollama pull llama3.1 or create a new agent with a custom model (https://github.com/i-am-bee/bee-stack?tab=readme-ov-file#custom-models).

[matiasmolinas]

Hi @Tomas2D, I tried again with OpenAI and Ollama on two different machines, also removing all the Docker images to ensure I am using the latest version. However, I am still experiencing the same two issues.

[Tomas2D]

Could you send me the logs?

./bee-stack.sh logs

[matiasmolinas]

2025-01-25_1124S.zip Hi @Tomas2D,

I’ve attached the logs as requested. Please let me know if there’s anything on my side that might be causing the issue. Thank you for your help!

[Tomas2D]

Log files are empty. cc @jezekra1

[jezekra1]

Did you collect the logs after ./bee-stack.sh start? The logs are collected only from running containers.
getDefaultAgent error was a know issue, are you sure you are on the latest version? If so, we should release a new version of bee-api which should have the fix.

Ollama should be working though, is your ollama server running on 0.0.0.0 as described in the setup script? Did the setup check for ollama pass?

[matiasmolinas]

Hi, thank you for the detailed feedback! I performed a fresh install, and now Ollama is working perfectly 🎉.

As for the OpenAI adapter, could you let me know if it's ready for testing? I'd love to give it a try and help validate it if possible.

[xjacka]

OpenAI should work.

[MohHeader]

I have the same issue with a fresh new setup

 git clone https://github.com/i-am-bee/bee-stack.git
cd bee-stack
./bee-stack.sh setup

Choose LLM provider:
[1]: watsonx
[2]: ollama
[3]: openai
Select [1-3]: 3  

Provide OPENAI_API_KEY: sk-proj-***
Do you want to enable docling text extraction? ⚠️ Requires >= 15GB of RAM **CONFIGURED** for the container runtime ⚠️ (Y/n): n
Do you want to start bee-stack now? (Y/n): Y

Then I get that error

[Tomas2D]

I was able to reproduce the problem, we will soon provide a fix. Thanks for the report.

[xjacka]

This is probably a bug with the instrumentation of the openai library.

[matiasmolinas]

Is it in the plans to fix it and add other providers? Is it safe to try contributing by adding or using another provider?

[benmoumen]

Hello and Thank you for the amazing platform.
I'm getting the same issue with OpenAI.
Logs provided:
2025-02-19_1527S.zip

[Tomas2D]

There is no plan to fix it as the bee-stack is no longer maintained.

We are now working on BeeAI Platform (Bee Agent Framework) and Bee AI Platform ()