title: TinyLLaMA Chatbot
emoji: 🤖
colorFrom: indigo
colorTo: purple
sdk: docker
app_file: main.py
pinned: false
🤖 TinyLLaMA 1.1B Chatbot (FastAPI + LoRA)
An advanced conversational‑AI chatbot powered by TinyLLaMA‑1.1B‑Chat‑v0.6, fine‑tuned with LoRA (Low‑Rank Adaptation) and served through FastAPI.
This Space uses the Docker SDK, so everything needed to run is contained in the supplied Dockerfile.
🧠 Model Details
🚀 Key Features
- Real‑time Q&A with natural‑language prompts
- FastAPI backend for low‑latency inference
- LoRA fine‑tuning keeps VRAM and disk usage small
- Clean responsive HTML/CSS frontend (Jinja2 templates)
- Fully containerised for reproducible deployment (Docker + Hugging Face Spaces)
💻 Tech Stack
| Layer | Tech |
|---|
| Backend | FastAPI, Python 3.10+ |
| Model I/O | Hugging Face Transformers, PEFT |
| Frontend | Jinja2, Vanilla CSS |
| Container | Docker |
| Platform | Hugging Face Spaces (Docker SDK) |
📦 Local Setup (non‑Docker)
git clone https://huggingface.co/spaces/JigneshPrajapati18/Chatbot
cd Chatbot
python -m venv venv
source venv/bin/activate # Windows: venv\Scripts\activate
pip install -r requirements.txt
uvicorn main:app --host 0.0.0.0 --port 7860
