gfinn — AI Lending Chatbot
AI assistant for lending company. Fine-tuned LLM with QLoRA on NVIDIA DGX Spark for automated loan inquiry responses.
PythonFastAPIOllamaQLoRADGX Spark
Gallery
gfinn is an AI Chatbot designed for a lending company, capable of automatically answering questions about loan products, terms, interest rates, and application processes — reducing Call Center workload.
Built using QLoRA fine-tuning (4-bit quantization) on NVIDIA DGX Spark with 121GB unified memory, ensuring the model accurately understands Thai lending business context.
The backend is built with FastAPI connected to Ollama for LLM inference, supporting multi-turn conversations and direct database queries.
Key Features
Fine-tuned LLM for Lending Business
QLoRA 4-bit Training on DGX Spark
Multi-turn Conversation
Loan Database Integration
FastAPI + Ollama Backend
100% Thai Language Support
Interested in a project like this?
Free consultation — tell us what you need and we will help you build it.
Start Your Project
