LocalLLM Chat Interface
Local AI chat application with custom agent personas, knowledge base upload, and web search — all running privately on your machine.
December 1, 2025
LLMsFastAPIPythonAI Agents
Overview
A personal project building a fully local AI chat command center. Connects to LM Studio for running large language models privately on your own hardware, with a web-based UI for conversations.
Features
- Persistent Chats: Chat history saved across sessions
- Custom Agent Personas: Define AI agents with custom system prompts for different tasks (coding assistant, writing helper, researcher, etc.)
- Knowledge Base: Upload PDFs to give the AI context about your documents — text is extracted and used for informed responses
- File Attachments: Attach files to conversations for analysis
- Web Search: DuckDuckGo integration for real-time web lookups during conversations
- Privacy: Everything runs locally — no data leaves your machine
Architecture
- Backend: FastAPI server connecting to LM Studio's OpenAI-compatible API
- Frontend: Clean web UI (HTML/CSS/JS)
- Document Processing: pypdf for PDF text extraction into the knowledge base
Technologies
Python, FastAPI, OpenAI API (LM Studio compatible), HTML/CSS/JS, DuckDuckGo Search, pypdf