一个轻量级、支持全链路且易于二次开发的大模型应用项目(Large Model Data Assistant) 支持DeepSeek/Qwen3等大模型 基于 Dify 、LangChain/LangGraph、Ollama&Vllm、Sanic 和 Text2SQL 📊 等技术构建的一站式大模型应用开发项目，采用 Vue3、TypeScript 和 Vite 5 打造现代UI。它支持通过 ECharts 📈 实现基于大模型的数据图形化问答，具备处理 CSV 文件 📂 表格问答的能力。同时，能方便对接第三方开源 RAG 系统 检索系统 🌐等，以支持广泛的通用知识问答。

程序员鱼皮的 AI 知识库，汇总热门的 AI 大模型和工具，比如 Deepseek 使用指南、提示词技巧、知识干货、应用场景、AI 变现、行业资讯、教程资源等一系列内容，帮助你快速掌握 AI 技术，走在时代前沿。涉及大模型 ChatGPT、Claude、Gemini、Deepseek、Deepseek-r1、QWEN、GROK 等；涉及技术 Spring AI、LangChain、RAG、MCP、A2A 等；涉及 Cursor 等工具。本项目为开源文档版本，网站已升级为编程导航 AI 板块

Model Context Protocol server for DeepSeek's advanced language models

MCP Server for AI Summarization

This lightweight Chrome extension lets you pin and manage your most important Deepseek chats for improved productivity, keeping them easily accessible at the top of your sidebar so you never have to scroll through your chat history again.

Code for benchmarking the speed of DeepSeek R1 from different providers' APIs.

Real-time transcription, AI-driven answer suggestions, and interview simulation using Next.js, React, Azure Speech, Gemini and OpenAI. Supercharge your interview skills! Check this for Demo. This is similar to finalround AI which is costly

Ollama Optivus is a video processing AI that extracts meaningful insights from videos using AI technologies like YOLO for object detection, EasyOCR for text extraction, and Whisper for speech-to-text. Integrated with Ollama LLM, it provides detailed summaries and event analysis, supporting various advanced AI model prompting.

Devseek.ai is an AI-powered SaaS platform that completely automates the software development lifecycle (SDLC)—from ideation to deployment—without requiring human developers. It replaces software engineers with AI agents that write code, test applications, fix bugs, and generate documentation, delivering a fully functional, production-ready app.

This is WebUI (Frontend, server for API/generate) based on Ollama serve to interact with DeepSeek R1 via browser

A client compatible with OpenAI API that supports displaying thought chains - DeepSeek-R1, OpenAI o3/o4-mini etc.

Chat summarizer using deepseek:r1, ollama and whatsapp-web.js

This is my portfolio developed with React, Vite, CSS, and JavaScript. The goal of this project is to showcase my skills, projects, and experiences in an interactive and responsive way.

AI-powered coding assistant with DeepSeek integration for VS Code, featuring intelligent code completion, chat assistance, and documentation generation.

Retrieval-Augmented Generation (RAG) combines information retrieval with AI-generated responses to improve accuracy and contextual relevance. This project demonstrates the design and implementation of a RAG-based system using Node.js, Express, LangChain, and MySQL, optimized with caching, parallel processing, and AI-driven query handling.

User-friendly LLM chat application that works with the OpenRouter API - currently only works with predefined free models

DeepSeek-R1: WebGPU-based Local Reasoning Model

3D Mentor is an experimental AI-driven educational interface that combines Three.js, LLMs, and speech technologies to create a lifelike 3D mentor who can understand and respond to users in real time. It aims to transform digital learning by providing a more immersive, interactive, and human-like experience.

DeepSeek to PDF: Chrome and Firefox Extension for Easy Chat-to-PDF Conversion

This project uses React + Vite to create a virtual assistant based on the Deepseek R1 API, allowing visitors to obtain information about my profile in an interactive and dynamic way.