← sources

Local vision model for 3D print monitoring

source: reddit (reddit:LocalLLaMA)published 2026-09-26 04:40ingested 2026-09-27 02:40

retained text

I'm thinking a small model that would look at camera input while printing and detect obvious failed prints, spaghetti, bed adhesion problems, things of that nature. That way it could notify the user in the case of catastrophic failure, saving filament and equipment, maybe even useful for fire safety.

Anyone try something like this or have ideas on how to implement? What kind of model size might be feasible?

  
EDIT: I have \~42GB to work with 

document fetched 2026-09-27 00:39 · canonical https://reddit.com/r/LocalLLaMA/comments/1wqheo9/local_vision_model_for_3d_print_monitoring