Is your feature request related to a problem? Please describe.
ChatGPT is a machine learning model that receives a text/image prompt and responds with conversational text.
There are some rudimentary ML models that generate 3D models based on a text prompt, e.g. NVIDIA GET3D.
CPQ is approaching the milestone where a user can drag-drop an STL and it slices and prints to completion on a farm of one or more printers (see #81).
So, how about this:
A user writes text describing the object they want, and hits "submit"
A 3D object is generated, queued, printed automatically
Describe the solution you'd like
It doesn't have to be perfect - this would be more of a demo showing off the concept. The ML execution would best be as a separate server, and a separate OctoPrint plugin would handle calling out and receiving a file back. This will likely look very similar to PrePrintService.
Describe alternatives you've considered
There may be an opportunity for a meta-prompt, i.e. the user describes a problem they have and ChatGPT comes up with a list of suggested 3D parts that could address it, that the user then picks from.
Is your feature request related to a problem? Please describe.
ChatGPT is a machine learning model that receives a text/image prompt and responds with conversational text.
There are some rudimentary ML models that generate 3D models based on a text prompt, e.g. NVIDIA GET3D.
CPQ is approaching the milestone where a user can drag-drop an STL and it slices and prints to completion on a farm of one or more printers (see #81).
So, how about this:
Describe the solution you'd like
It doesn't have to be perfect - this would be more of a demo showing off the concept. The ML execution would best be as a separate server, and a separate OctoPrint plugin would handle calling out and receiving a file back. This will likely look very similar to PrePrintService.
Describe alternatives you've considered
There may be an opportunity for a meta-prompt, i.e. the user describes a problem they have and ChatGPT comes up with a list of suggested 3D parts that could address it, that the user then picks from.
Additional context
See links in description