This project provides a FastAPI wrapper for interacting with Google Bard, a conversational AI by Google. It allows users to send messages to Google Bard and receive responses through a simple API.
-
Updated
Apr 3, 2024 - Python
This project provides a FastAPI wrapper for interacting with Google Bard, a conversational AI by Google. It allows users to send messages to Google Bard and receive responses through a simple API.
A multimodal chat interface with many tools.
Lightweight MCP client that uses a local Ollama LLM to query multiple MCP servers defined in config.json
🦙 chat-o-llama: A lightweight, modern web interface for AI conversations with support for both Ollama and llama.cpp backends. Features persistent conversation management, real-time backend switching, intelligent context compression, and a clean responsive UI.
💻🔒 A local-first full-stack app to analyze medical PDFs with an AI model (Apollo2-2B), ensuring privacy & patient-friendly insights — no external APIs or cloud involved.
A lightweight, intuitive templating language designed for interactive use in LLM chat sessions.
Simple MLX + Streamlit Chatbot
💻🔒 A local-first full-stack app to analyze medical PDFs with an AI model (Apollo2-2B), ensuring privacy & patient-friendly insights — no external APIs or cloud involved.
Educational AI chat client: provider abstraction, token compression & state management in ~600 lines Python. Learn robust AI integration patterns.
A stylish terminal-based DeepSeek AI chatbot built with Python featuring multiple model selection, rich UI panels, and real-time chat responses.
A Docker-based chat interface that communicates locally with an Ollama instance using React (Vite) and a FastAPI backend. Ideal for quick experiments with LLMs on your own computer.
A two-way session bridge between VS Code Copilot Chat and external automation. / 外部自动化与 VS Code Copilot Chat 之间的双向会话桥。
Lightweight multimodal AI chat interface powered by Claude Code
🏥 AI Healthcare Assistant – An interactive AI-powered medical chatbot built with Streamlit, offering conversational healthcare info, voice input, document support, offline knowledge, and emergency keyword detection.
A Django-based RAG knowledge base demo with local LLM generation, asynchronous PDF ingestion, Pinecone vector retrieval, and a template-based chat interface.
A secure, AI-powered conversational interface to develop, debug, and operate code from your phone : replacing SSH, terminals, and fragmented workflows with a single chat-based system.
Web interface for Ollama local AI models with streaming responses, conversation management, model switching, and chat export functionality.
SmartBI is an AI-powered tool that lets users query databases using plain English and get instant insights without writing SQL.
AI assistant built to support hotel operations and guest services
Streamlit MCP Agent is a web application that provides a chat interface for AI agents using the Model Context Protocol (MCP). It integrates various tools via MCP clients to enable interactive AI assistance, suitable for corporate web app deployment.
To associate your repository with the chat-interface topic, visit your repo's landing page and select "manage topics."