Local Multi AI Models API

---

How to Run

Run the runner.py script.
This will execute the command:
uvicorn.run("api:app", host="0.0.0.0", port=8000, reload=True)
and start a server on port 8000.

The API uses FastAPI. This project is a work in progress.

---

Available Classes

1) custom_model

This class imports and uses a pre-installed model from Hugging Face.
It provides the following endpoints:

• POST:
  • /initModel – loads the model into memory.
  • /delModel – frees the model from memory.
• GET:
  • /getResponse – returns a string with the model's response.

2) generator3D

Generates a 3D object from text or image using Shap-E.

3) openrouter_deepseek

Calls the DeepSeek free model using OpenRouter.

---

Status

This project is under active development. The code may change, so this README might refer to an older version.
All dependencies are listed in the requirements.txt file.