Local GPT Website Chatbot Deployment

I am ready to run a GPT-style large language model directly on my own machine so I can serve a real-time chatbot embedded on my website. I have the hardware available but need an expert who can install the model, configure all dependencies, and expose an endpoint that my front-end widget can call.

Here is what I have in mind:
• Select and download an open-weight GPT-like model that can reasonably run on local hardware (e.g., Llama-2, Mistral, or another suitable alternative).
• Set up the execution environment—Python, CUDA, PyTorch or TensorFlow—plus any supporting libraries (LangChain, FastAPI, uvicorn, etc.).
• Create or refine an inference script that keeps response times low enough for smooth chat.
• Build a lightweight API (REST or WebSocket) so the website can pass the user’s prompt and receive the model’s reply.
• Hand me clear, repeatable launch instructions (ideally Docker-based) so I can restart or migrate the service without hassle.

Acceptance criteria
1. Model loads and answers a sample prompt locally with no external calls.
2. My website can hit the exposed endpoint and display replies in the chat widget.
3. All setup steps are documented in a concise README or shell script.

If you have experience squeezing the best speed/quality trade-off out of local GPT models and can walk me through any GPU driver quirks, let’s get started.