Did you tried this with ollama local models?

[drfenixion]

Hello. Greate work!
I want to integrate this into RobotCAD but with local models and as an option with api providers choosing.

Did you tried this with ollama local models like Qwen3 if so that results? If - "no" do you have any plans to do?

[yas1nsyed]

Hi! Thanks for the appreciation and interest in integrating this with RobotCAD!

I’ve tried running it with Ollama local models, but the results weren’t very good. To get better performance, we need to either use a RAG with the LLM or use a fine-tuned model.
The API providers option can likely be integrated though.