Vector dimension 1536 does not match the dimension of the index 512 issue

[Vishnu_Subramoniam]

Describe the problem/error/question

I am using the same models (text embedding open ai-small) in pinecone and open-ai embedding but I am getting the following error: Vector dimension 1536 does not match the dimension of the index 512. Can someone help me fix this, thanks!

What is the error message (if any)?

Vector dimension 1536 does not match the dimension of the index 512.

Please share your workflow

Share the output returned by the last node

Pinecone node keeps running and then stops with the above error message

Information on your n8n setup

• **n8n version: 1.104.2
• **Database (default: SQLite): Default
• n8n EXECUTIONS_PROCESS setting (default: own, main):
• **Running n8n via (Docker, npm, n8n cloud, desktop app): n8n CLoud
• **Operating system: Mac OS

[jabbson]

Hey @Vishnu_Subramoniam hope all is well. Welcome to the community,

This means that your vector dimensionality does not match between the embedding model and the database. Try changing the dimensions in the options:

[2020_qadan]

Hello, I’ve also encountered the same issue. However, I’m using the Ollama model, not Openai, and there’s no option to change the dimension. How should I handle this problem?

[jabbson]

Hey @2020_qadan hope all is good.

Please create a separate topic for your question, I will be more than happy to discuss how this can be done with Ollama.