Top Open Source AI Dev Influencers in April 2025

Implementing DeepSeek R1's GRPO algorithm from scratch

A cutting-edge voice assistant combining real-time speech recognition, AI reasoning, and neural text-to-speech capabilities.

Official PyTorch implementation of the paper "Chapter-Llama: Efficient Chaptering in Hour-Long Videos with LLMs"

[Support 0.48.x]（Reset Cursor AI MachineID & Auto Sign Up / In & Bypass Higher Token Limit）自动注册 Cursor Ai ，自动重置机器ID ， 免费升级使用Pro功能: You've reached your trial request limit. / Too many free trial accounts used on this machine. Please upgrade to pro. We have this limit in place to prevent abuse. Please let us know if you believe this is a mistake.

Used for thinking process intervention of reasoning models such as DeepSeek-R1, effectively controlling the reasoning thinking process. 用于DeepSeek-R1等推理模型的思维过程干预，有效控制推理思考过程

OmniMind: An open-source Python library for effortless MCP (Model Context Protocol) integration, AI Agents, AI workflows, and AI Automations. Plug & Play AI Tools for MCP Servers and Clients, powered by Google Gemini.

Enable tool-use ability for any LLM model (DeepSeek V3/R1, etc.)

在Mac上记录您在任何应用打下的字与您剪切板的内容并在本地通过ollama DeepSeek-r1分析。Reminding all the words you have typed for Macbook and analysis with ollama Deepseek-r1

An AI-Powered Multi-Agent Solution for Retail Banking Products.

code editor with completion powered by llamacpp

⽊ ongoing list of curated resources for decentralized AI and LLMs, with a focus on agents, blockchains, and privacy 🏴‍☠️

Simplified "how to run" Llama 4

eShopLite is a set of reference .NET applications implementing an eCommerce site with features like Semantic Search, MCP, Reasoning models and more.

A conversational AI agent built with Python, Streamlit, and LangChain, featuring multi-LLM support (Gemini, Llama 3), persistent chat history via PostgreSQL, agent tools (Web Search, Python REPL), and image generation using Google Cloud's Imagen model.

An intelligent web application that transforms complex questions into interactive visual mind maps using AI (Deepseek integration).

An interactive Streamlit app that lets users upload PDFs and ask questions to receive concise, reasoning-based answers using the DeepSeek-R1 language model.

A fine-tuned version of SmolLM2-360M-Instruct-bnb-4bit specialized for parsing unstructured calendar event requests into structured JSON data.

A fast, lightweight CLI tool for versioning, testing, and optimizing your AI prompts across multiple providers. Easily track prompt evolution, run A/B tests, and measure performance without Git dependencies. Supports OpenAI, Claude, Llama, and HuggingFace.

Run RamaLama as part of your Llama Stack

A powerful Android app that connects to LM Studio running on your computer, allowing you to chat with your favorite AI models from your mobile device.

Fine-tuned the Gemma 2B-IT model to generate concise summaries of complex legal documents. Leveraged a custom dataset with legal instructions, gold summaries, and system outputs to train the model for efficient legal text comprehension.

Multi-Agent-AI-Researcher-Powered-by-DeepSeek-R1-main

ChatGPT 中文版：免费直连指南（支持GPT-4、4o、o1、o3 和 DeepSeek R1，无需翻墙）

ChatGPT 中文版：国内直连指南（支持GPT-4、GPT-4o、o1、o3 和 DeepSeek R1，无需翻墙）【2025年4月 更新】

LlamaResearcher website

Front for La Muralla project

Fork of lm/navigation-stack-backport

Using Regex,Bert,Linear_regressin,Llama

Analysis of Text data from Twitter that surfaced during the launch of DeepSeek R1 in January 2025. Spoiler Alert! Tweets with negative sentiment got more engagement

An interative chatbot for overall creativity and productivity, based on DeepSeek R1 API.

This repository focuses on the fine-tunning of the Code-Assistant LM, that uses Phi-2 as it's base. Uses LoRA to learn coding leetcode questions.

An AI-driven tool that converts voice input into professional emails using Whisper for speech-to-text and LLaMA-3 for email refinement. Includes automated formatting and SMTP integration for direct email sending.

Simple chat bot based on deepseek r1

On-Device Multi-Turn Chatbot Powered by Llama 3.2 1B

a management webui for llama.cpp

Innovative tool that leverages advanced GenAI techniques with the Llama model to generate engaging LinkedIn posts. It processes LinkedIn content and uses a few-shot learning approach to create posts tailored to your selected preferences - whether you need a professional update, an inspirational story, or a casual share.

A cutting-edge Cascading voice assistant combining real-time speech recognition, AI reasoning, and neural text-to-speech capabilities.

Basic AI agent which can help with appointment booking and answer basic queries

Fire-Detection-Siglip2 is an image classification vision-language encoder model fine-tuned from google/siglip2-base-patch16-224 for a single-label classification task. It is designed to detect fire, smoke, or normal conditions using the SiglipForImageClassification architecture.

Medical Chatbot

公众号接入LLM（deepseek r1、v3...），实现AI聊天问答功能。扩展功能：联网功能和调用部分工具功能，如查询天气、新闻日报等

ChatGPT 中文版：免费直连指南（支持 GPT-4o、GPT-4、o1、o3 和 DeepSeek R1、Claude3.7，国内无需翻墙）【2025年4月最新】

ChatGPT 中文版：国内直连指南（支持GPT-4、4o、o1、o3 和 Claude3.7、Grok3、DeepSeek R1，无需翻墙）【4月最新更新】

The LMS project built with Payload CMS

library management system

Web de practica para lenguaje de marcas y CSS

Basic .NET chat app

Optimized Trainable Implementation of Llama 3.2 1B for multi-GPU setups. Suitable for max-likelihood evaluation, pre-training, and fine-tuning.

利用deepseek-r1：14b模型加上rag和langchain实现的简易校园问答助手

Build your own QA System locally based on FastGPT and DeepSeek R1

Learning management system like udemy using react js

A modern web chatbot powered by GROQ API, built with React and Flask. Features real-time AI responses, authentication, dark mode, and chat history persistence. Topics: chatbot, react, flask, groq, llama, ai, python, javascript

AI Data Analyst Web App with DeepSeek R1 chatbot

NutriScan is a software that uses the LLaMA-3.3-70B-Versatile model to scan food labels and images, giving easy-to-understand nutrition and health advice.

Personal project for creating a chat room with AIs using LM Studio as a backend

this is a node js Learning management system which is actually in development right now stay tuned for great features and new approaches to solve complex problems

LLama_lora_note

This is a web-based application that uses an AI model (via Ollama and LLaMA 3) to analyze and grade English essays. Upload a .txt essay, choose grading preferences, and receive inline feedback with rubric-based scoring, strengths, weaknesses, and suggestions. You can even download a PDF report of the graded essay!

中文石油工程领域DeepSeek-R1蒸馏数据集

DeepSeek-R1 免费的API可使用。https://hao666666.online 获取免费API接口。

CS 4650 Final Project

This is a hybrid app that stream from my school's website

Fine-tuned Meta-Llama 3.1–8B on a curated dataset that improve code reasoning and explanation quality

OOs modelos de Linguagem de Grande Escala (LLMs), como GPT, LLaMA e BERT, revolucionaram o processamento de linguagem natural (NLP), oferecendo desempenho excepcional em tarefas como geração de texto, classificação e tradução.

This notebook fetches content from a webpage, removes irrelevant sections

Analyze Stocks using Deepseek-R1 and Ollama

Adapted the DeepSeek model to a medical dataset to improve disease prediction accuracy through fine-tuning.

A chatbot assistant for businesses, using Python (FastAPI), LLama-Index, and the RAG technique to specialize models.

Struggling to write cold emails that get replies? This AI tool crafts personalized, spam-safe emails with high-open subject lines in seconds. Perfect for sales, recruiters, and founders. Boost responses—try it today!

Comparative study of LoRA and diagonal-masked PEFT methods on LLaMA 3.1 using Korean instruction datasets.

Let you quit 100+ servers discord in one click !

A unified web application integrating multiple AI chatbots like ChatGPT, Claude, Gemini, Llama, and more. Users can interact with each model independently within a single interface.

Landing page about Free-GPT4-WEB-API

This project is a Retrieval-Augmented Generation (RAG) chatbot built with Streamlit, using the DeepSeek R1 Distill LLaMA 70B model via Groq API. It allows users to upload documents (PDF, TXT, DOC, DOCX), extract text, and ask questions based on the document content.