Fast and Effective German Summarisation service with a Custom Large Language Transformer Backend
Deployed Demo: zusammenfassen.vercel.app
Note, there might be initial latency due to HuggingFace loading the model.
- HuggingFace Transformers
- PyTorch
- NextJs
- TailwindCSS
- DaisyUI
- FastAPI
- A t5-small base transformer was fine-tuned on the german subset of the MLSUM dataset
- To run the NextJs frontend
cd frontend
npm install
npm run dev
- To run the FastAPI backend
cd backend
pip install -r requirements.txt
python main.py
- Add Backend Endpoint to
.env.local