This is the official repository for "Towards Foundational AI Models in Additive Manufacturing: Evaluating Large Language Models for G-Code Debugging, Manipulation, and Comprehension". This repo contains all results that were generated as part of our experimentation - G-code, markdown file for chats, chat screenshots, and an overview of our prompts.
Results are split into two categories - raw and processed.
raw contains raw output from the model, and are not processed in any way. These results are stored in the raw folder.
Outputs for more models soon!
Notes:
- There maybe spelling errors in some input prompts.
- We highlight "User" and "Model name" for easy viewing.
- We add backticks to all G-codes (input) for easy viewing. Backticks in outputs are generated by respective models.
- In some cases, missing backticks were added to the end of the ChatGPT responses to make it easier to read.
- Minor differences in codeblock formatting are generated by the model, and are not manually added.