Batteries-included multi-agent framework. Agents for databses, files, emails, APIs and web crawler. Self-host, run on your data. Multi-agent, multi-model workflows. Access from Slack, Telegram, desktop app or web chat.

Cross-platform desktop app for multi-model AI chat, web search, knowledge bases, and document Q&A — install and use out of the box. 跨平台本地客户端，集成多模型聊天、网络检索、知识库与文档对话，开箱即用、稳定高效。

A open source multi-model chat app. Chat with local or cloud models . Desktop App . Web App . with agent support

A self-hosted, production-grade web application that leverages local speech-to-text (Whisper) and text-to-speech (Coqui TTS) models, along with a Hugging Face-powered conversational AI and Qdrant-based long-term memory for natural, context-aware voice interactions. The app features multi-user authentication, real-time chat with WebSocket support.

Multi-provider AI chat client supporting OpenAI, Gemini, OpenRouter, and custom LLMs with reasoning model support, multimodal input, and conversation management. Desktop (Electron) & web app.

A zero-dependency, single-file Progressive Web App (PWA) for OpenRouter. Features local IndexedDB storage, multi-threading, dynamic per-chat model switching, and 100% client-side privacy.

Portfolio project showcasing a simple multi-model chatbot. Chat with LLMs from OpenAI (GPT), Anthropic (Claude), DeepSeek, and Google (Gemini). Includes both a terminal interface and a Streamlit web app, with options for custom bots and local chat history.

A chatbot web app built with Streamlit, powered by Ollama's LLaMA model for AI responses. It uses FAISS for efficient similarity search and SQLite for chat history storage. The app supports multi-turn conversations, caching, and an interactive UI for managing chat history. Let me know if you want a more customized description! 🚀

The Multi-Function Chat App is a versatile web application built using Streamlit, integrating advanced AI models for various functionalities. It allows users to interact with text, PDFs, websites, and images in multiple ways, such as language translation, summarization, and content analysis.

This repository contains a user-friendly web application for chatting with advanced AI language models, including GPT-4, GPT-3.5, and Llama 2, all accessible in a single interface. Built using Python and Streamlit, the app is designed to be clean, intuitive, and accessible for users of all skill levels.

A multi-model chatbot web app is a single platform that lets you switch between different AI models (like Deep Seek, Meta lemma, or Mistral). This allows you to pick the best AI for tasks like writing, coding, or getting information, all from one user-friendly chat interface.

Multi model ai chat web app

Authenticated AI desktop app with multi-model chat, web search, Better Auth, and a Next.js/Postgres backend.

A multi-agent financial assistant built with FastAPI, Groq’s Llama-3 model, and Phidata , combining real-time stock insights and web search in one chat app.

Multi Llama Chat is a fast, client-side web app to compare multiple Ollama models side-by-side in one workspace. Tag models with @, assign custom roles (tester, designer, reviewer, etc.), run inter-model discussions, duplicate model instances, and offline support.

A production-grade AI assistant web app powered by Google Gemini. Features multi-turn conversation, model selection (Flash/Pro), customizable system prompts, temperature control, chat export, and a custom dark-theme UI. Built with Python + Streamlit.

A production-grade AI assistant web app powered by Google Gemini. Features multi-turn conversation, model selection (Flash/Pro), customizable system prompts, temperature control, chat export, and a custom dark-theme UI. Built with Python + Streamlit.

This project is a multi-session chatbot web app built with Python, NiceGUI, and OpenAI's GPT models. It features persistent chat history, AI-generated chat titles, markdown support, and an intuitive interface for managing and revisiting conversations. Ideal for learning, prototyping, and experimenting with conversational AI.

Hyperchats is a lightweight, TypeScript based conversational AI app built in six hours for a rapid-build challenge. It runs a transformer model on Cloudflare’s edge for fast, low-latency responses and supports smooth, multi-turn chats through a simple, modern web interface.

The Velin Assistant is a powerful, single-page web application designed for seamless, multi-functional interaction with a large language model (LLM). Built using vanilla HTML, Tailwind CSS, and JavaScript, it provides a fast and responsive chat experience that feels like a native app.

Built an AI-powered web app to classify multi-turn customer chats and generate rationales in real time. Used a hybrid model with DistilBERT for intent extraction and Gemma 2B for generative reasoning. Deployed via Flask with batch processing, live predictions, and instant PDF report generation.

A multi-functional AI web application built with React that allows users to chat with an AI, upload images for analysis, and generate high-quality images using AI models like Gemini . The app features an intuitive interface, real-time responses, and seamless integration of AI-powered functionalities.

Image Conversation Chatbot lets you chat about images. Upload a photo or paste a URL, and the app uses a vision‑language model to analyze content, answer questions, and extract details across multi‑turn dialogue. Built with Python, it provides a simple web UI for interactive, context‑aware image understanding.