Support for Mistral OCR via Custom Nodes

Feature Request Node Improvement
1

I’m trying to create a custom node that leverages Mistral’s Document OCR model, but I’m not able to do so. I’d like to use Mistral’s because From what research I’ve done, it’s the best-in-class model for this.

Here’s what I see when I’m trying to create the custom node:

image
image1034×932 80.7 KB

This is when I pasted the Python from their documentation site here: Basic OCR | Mistral AI

I’m trying to analyze construction drawing documents, hence the need for SOTA pdf reading models

2

Hey @samhbarton – I cut a ticket to add Mistral packages to our dev environment so you can use those, I’ll keep you posted.

3

Legend ty!

4

Hey @samhbarton – We added the packages for Mistral on the Run Code and Custom Node environments, let me know if it works as expected now :slight_smile:

5

This topic was automatically closed 3 days after the last reply. New replies are no longer allowed.