Skip to content
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

Error in the AI ​​agent node. #11515

Open
rovictorbh opened this issue Nov 2, 2024 · 4 comments
Open

Error in the AI ​​agent node. #11515

rovictorbh opened this issue Nov 2, 2024 · 4 comments
Labels
in linear Issue or PR has been created in Linear for internal review

Comments

@rovictorbh
Copy link

Bug Description

When the node is executed it changes the open ai model to gpt 3.5 and generates an error.

To Reproduce

  1. Add an agent to the workflow
  2. Choose a model from open ai. I use the 4th mini.
  3. Add instructions and tools.
  4. Run the workflow.
  5. The error is displayed.
  6. The reason is that it changed the model to gpt 3.5 during execution.
  7. Since the instructions and tools are prepared for the complexity of the model (4th mini), the error is displayed.

Expected behavior

The expectation is to run the execution normally. The workflow was working normally until the last update.

Operating System

Ubuntu 20.04.6 LTS (Focal Fossa)

n8n Version

1.65.2

Node.js Version

v18.20.4

Database

PostgreSQL

Execution mode

main (default)

@Joffcom
Copy link
Member

Joffcom commented Nov 2, 2024

Hey @rovictorbh,

We have created an internal ticket to look into this which we will be tracking as "GHC-379"

@Joffcom Joffcom added the in linear Issue or PR has been created in Linear for internal review label Nov 2, 2024
@OlegIvaniv
Copy link
Contributor

Hi, @rovictorbh,

Could you please share more details about this issue? I couldn't reproduce it. How did you determine that the model changed to GPT3? Some screenshots or a workflow JSON would be a great help.

Thanks!

@rovictorbh
Copy link
Author

rovictorbh commented Nov 6, 2024 via email

@OlegIvaniv
Copy link
Contributor

Thanks, @rovictorbh, and the execution you're showing is recent; on the video, it's from 2.11. Couldn't it be that you've executed this workflow before with gpt-3.5-turbo? It seems to me that it's a visual-only issue; on the backend, the selected model should be used. Can you confirm, by sharing part of that workflow, specifically agent and connected Open AI Chat Node? Also, in the input panel of the Open AI Chat node, there's information about which model was used; what does it say there?
CleanShot 2024-11-06 at 15 09 36

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
in linear Issue or PR has been created in Linear for internal review
Projects
None yet
Development

No branches or pull requests

3 participants