Top Open Source AI Dev Influencers in May 2025

a GitHub repository dedicated to building and hosting wheels for llama-cpp-python. The goal is to automate the build process for various platforms, architectures, and backend configurations using GitHub Actions and host the resulting wheels, likely via GitHub Releases or GitHub Pages.

Investigating formal testing of LM32 CPU

A proxy written in Python to translate the Responses API requests sent by the OpenAI Conversation plugin for home assistant and send them to Llama.cpp or similar Completions API compatible servers.

showcasing behavioral anomaly detection for fraud prevention utilizing LLaMA-2-7B

This project fine-tunes the DeepSeek-R1-Distill-Llama-8B model using Unsloth on Google Colab to act as an advanced Medical Expert Assistant. It allows for precise reasoning and treatment recommendations using clinical knowledge.

ChatGPT中文版：国内最易懂的访问指南（支持GPT-4、4o、o1、o3 和 DeepSeek R1，无需翻墙）【5月最新更新】

ChatGPT中文版：国内访问指南（支持GPT-4、4o、o1、o3 和 DeepSeek R1，无需翻墙）【5月持续更新】

Web server implementation of Llama

swa2025050108lm1eznsplcxsv1fgf50gi2

Simple Library Management System using Spring Boot

Collection of SimHub tools for LMU

SEM 2 REACT project

Learning material system for Software Engineering student at FPT University Cantho Campus

Saya Voice Assistant for Discord AI voice bot: listens, detects keywords, chats via LM Studio, and replies with TTS or voice cloning.

Learning Management System

Mockups and design assets for the SKILL BRIDGE platform addressing gaps in traditional Learning Management Systems

note for llama cpp use

"An intelligent airline customer service chatbot built with Python, FastAPI, and LLaMa-2-7B via OpenRouter API, featuring conversation memory systems, natural language understanding, and a responsive web interface using HTML/CSS/JavaScript and Jinja2 templates.

PowerShell wrapper for llama-server that simplifies the process of listing and running .gguf models

SightSpeak is an AI-powered assistive tool that combines YOLOv8 object detection with LLaMA 3.2 voice output to help visually impaired users. It detects nearby objects in real time and provides natural voice descriptions, enabling safer, more aware navigation of the environment.

Model Context Protocol using LLama

Jarvis voice enable LLaMA model

ChatDoctor is a medical chat model fine-tuned on a large language model Meta-AI (LLaMA) using medical domain knowledge.

using LLAMA.cpp based model with discord BOT

A self-contained offline web application that allows users to upload sample trajectory data and generate new realistic trajectories.

La malédiction des dieux

GitHub Pages Site

Run LLMs on Android

three users student, techer and Admin. using reactjs, expressjs, and css for styling.

Tarea 5 DAW

LLaMA3 Edu Chatbot: Fine-Tuning vs RAG on Lecture Transcripts

LLM clone of llama along with files of working models and datasets.

bot-llamacpp

Unified Efficient Fine-Tuning of 100+ LLMs & VLMs (ACL 2024)

A lightweight AI chatbot built with Spring Boot, GraphQL, Spring AI, and the LLaMA model via Ollama

TaxMate – RAG-based Tax Law Query Bot A semantic legal assistant powered by Retrieval-Augmented Generation (RAG) to simplify access to Indian tax law. It utilizes FAISS-based retrieval, Cohere reranking, and local LLaMA 2 inference to provide accurate tax-related answers through an interactive Streamlit UI.

Projects build using Llama index an framework to build llm application

El proyecto consiste en un prompt sencillo que generera un factura mediante ciertos criterios previamente configurados en el formato de salida.

Repository to Finetune LLama 3.1 8B on MedQA Dataset

oRAG é uma API de Chat construída com FastAPI que implementa Geração Aumentada por Recuperação (RAG). Ela utiliza Sentence Transformers para busca semântica em um conjunto de documentos (como um arquivo CSV) e se integra com Ollama (usando o modelo Llama 3.2) para gerar respostas conversacionais e contextualmente relevantes.

Repository for self try about running LLM on container using llama.cpp

A RAG-powered AI assistant that helps parents with baby care advice, combining document knowledge with real-time web information. Built with Streamlit, Google's Generative AI, and FAISS for efficient semantic search.

Easily deploy LLMs with Ansible. Uses Docker with llama.cpp or ollama. Secured with whitelisted IPs.