The power of Claude Code + [Gemini Pro / Flash / O3 / Grok / OpenRouter / Ollama / Custom Model / All Of The Above] working as one.

Better than Claude Code or Gemini CLI especially for Monorepos

一个高性能的OpenAI格式API多密钥轮询代理服务器，支持负载均衡，使用 Go 语言开发。A high-performance OpenAI-compatible API proxy server with multi-key rotation and load balancing, built with Go.

Most intuitive Stable Diffusion WebUI for AI artists, developers & researchers

A Curated Benchmark Repository for Medical Vision-Language Models

ComfyUI-OmniGen2 is now available in ComfyUI, OmniGen2 is a powerful and efficient unified multimodal model. Its architecture is composed of two key components: a 3B Vision-Language Model (VLM) and a 4B diffusion model.

Display giant ASCII-art logos with colorful gradients in your terminal — like Claude Code or Gemini CLI.

Interleaving Reasoning: Next-Generation Reasoning Systems for AGI

Machine Mental Imagery: Empower Multimodal Reasoning with Latent Visual Tokens (arXiv 2025)

Herobot is your 24/7 customer service assistant that helps you manage multi-channel customer conversations effortlessly.

🚀 EdgeCareer – AI-Powered Career Coach Full Stack AI Career Coach built with React 19 + Next.js 15, Tailwind CSS, NeonDB, Prisma, Clerk Authentication, Inngest, Gemini API, and Shadcn UI. A cutting-edge AI-driven career platform that provides personalized job recommendations, AI resume reviews, and real-time career insights to help users

Jailbreaks GPT, Sora, Claude, Gemini ,deepseek this prompt unlocks rage mode

The Operating System for Agents

This is the official repo for paper "StableMotion: Repurposing Diffusion-Based Image Priors for Motion Estimation"

🚀enhanced GRPO with more verifiable rewards and real-time evaluators

A Step-by-Step Implementation of Google Veo 3 Architecture from Scratch

极简高效、易于集成、灵活扩展、上下文管理强大、适合新手的 LLM 智能体开发框架

Autoregressive Semantic Visual Reconstruction Helps VLMs Understand Better

AI-powered development orchestration for Gemini CLI - adapted from Claude Code Flow by ruvnet

Professional Load Testing for Any OpenAI-Compatible LLM API

ComfyUI GLM Nodes: Seamlessly integrate ZhipuAI GLM models into ComfyUI for enhanced text generation and image-to-prompt capabilities. Features include advanced text chat (video prompt expansion) and GLM-4V powered image description.

A powerful audio generation extension for ComfyUI that integrates AudioX models for high-quality audio synthesis from text and video inputs.

部署于 CloudFlare Pages 的 AI 语音服务，使用 siliconflow 的语音转录模型 SenseVoiceSmall 和 openai 的 gpt-4o-mini-tts

🤖 A composable framework for building AI applications.

A simple, "Ollama-like" tool for managing and running GGUF language models from your terminal.

Data-driven multi-agent ecosystem for autonomous decision-making in GLM深远海船舶应用赛

TTPMapper is an AI-driven threat intelligence parser that converts unstructured reports whether from web URLs or PDF files into structured intelligence. Using the DeepSeek LLM, it extracts MITRE ATT&CK techniques, IOCs, threat actors, and generates contextual summaries.

A framework for training diffusion models with stable, self-consistent scores near the data distribution.

Library for text-to-text regression, applicable to any input string representation and allows pretraining and fine-tuning over multiple regression tasks.

C# SDK for Azure Sora

Simple LLM package for edge devices.

A Java Spring Boot web application that generates Sora videos using Azure OpenAI API. This application provides a user-friendly web interface for creating AI-generated videos with configurable video specifications including resolution and duration.

[ICCV 2025] Official code for paper: Beyond Text-Visual Attention: Exploiting Visual Cues for Effective Token Pruning in VLMs

A simple way to go from Transkribus to HuggingFace for VLM training

This repository offers a straightforward way to access Apple Intelligence Foundation Models in your React Native apps. 🌟 With easy installation and clear API methods, you can quickly check model availability and configure sessions for optimal performance. 🛠️

Chat with an LLM

A custom LoRA loader node for ComfyUI with advanced block-weighting controls for both SDXL and FLUX models. Features presets for common use-cases like 'Character' and 'Style', and a 'Custom' mode for fine-grained control over individual model blocks.

A FastAPI-based application that processes images using multiple Vision Language Models (VLMs) to answer to user query.

A simple GUI wrapper for stable-diffusion.cpp written using C and GTK 4.

unsloth vlm grpo

Cerno is an open-source tool that enables deep, multi-step research with autonomous AI agents. It offers clear insights into each reasoning step, allowing users to manage complex workflows effectively. 🐙🌟

ForkSilly的文档。ForkSilly：兼容sillytavern V2角色卡（png）、世界书、正则、预设的安卓应用

AstrBot 插件，用于检测和管理大型语言模型 (LLM) 的输出。它能帮助你识别和响应 LLM 生成的特定内容，确保其输出符合你的预期和安全标准。通过此插件，你可以更好地控制 LLM 的行为，避免不希望出现的信息。

OneSDK: A unified AI access SDK for edge devices, providing LLM capabilities (text/voice chat, image generation) and IoT device management with MQTT support, compatible with ESP32, Linux, macOS, and Windows platforms.

[ICCV 2025] official code repository of paper "LoRD-HOI: Zero-Shot HOI Detection with Low-Rank Decomposed VLM Feature Adaptation"

finding a self-preference vector in Llama-3.1-8B-Instruct residual stream activations

DeepSeek R1-zero轻量级复现

Connect Claude Code with Google's Gemini AI for seamless collaboration on coding tasks. Use this tool for quick code reviews and to brainstorm ideas effectively. 🐙💻

# Augment UserThis tool automates the registration process for the AugmentCode platform. It simplifies tasks like email verification and provides a clear progress display. 🚀🛠️

The Private AI Setup Dream Guide for Demos automates the installation of the software needed for a local private AI setup, utilizing AI models (LLMs and diffusion models) for use cases such as general assistance, business ideas, coding, image generation, systems administration, marketing, planning, and more.

Explore the Foundation Models Playgrounds repository to see how to engage with Apple's Foundation Models through Swift. Each playground offers clear examples, from simple chats to dynamic storytelling. 🛠️📚

Recently I have been researching emergent complexities through first principles reductionism of Langton's Ant in the hopes that they could potentially offer insights into the emergent intricacies of frontier large language models.

🎓 Simulador interativo desenvolvido como trabalho final da disciplina ACH2063 — Introdução à Administração e Economia para Computação (USP). O EconoSim permite que alunos tomem decisões fiscais e monetárias em tempo real e visualizem os impactos sobre o modelo IS-LM de forma dinâmica e colaborativa.

All-in-one local AI stack for Mistral-13B and Llama.cpp, with one-step CUDA wheel install, OpenAI-compatible API, and modern web dashboard. Switch between local and cloud chat, run on your own GPU, and deploy instantly—no API keys or paywalls. Designed for easy install, custom builds, and fast remote access. Enjoy!

Containerized Go service that uses LLMs (e.g., LLaMA 3.1 Instruct) to analyze backend logs via RabbitMQ and store structured insights in PostgreSQL

This is a Rag-LLM chatbot demo for tour guide in Da Nang

Magcache implementation for SDXL

GangGPT revolutionizes Grand Theft Auto V multiplayer roleplay by integrating advanced AI systems that create a living, breathing virtual world. Built on RAGE:MP with Azure OpenAI GPT-4o-mini, this project delivers procedurally-generated missions, intelligent NPCs with persistent memory, and dynamic faction warfare-transforming traditional roleplay

Laravel Gemini AI Translation Extractor scans your Laravel project for translation keys, uses Google Gemini AI for translations, and generates language files automatically—streamlining and accelerating your localization workflow.

Prompto – a web-browser extension + Next.js dashboard that intercepts your GPT/Claude prompts, auto-applies 25 advanced engineering tricks (CoT, XML guards, compression, role, etc.), shows a diff, and delivers higher-quality answers with optimized token usage.

Code for the experiments and websites of the paper "Same Task, Different Circuits"

ChatGPT 中文版：国内ChatGPT镜像网站整理合集（支持 GPT-4、GPT-4o、o1、o3及Claude 4 Sonnet、Gemini 2.5 Pro、Grok3，无需翻墙）【7月持续更新】ChatGPT 中文版是基于 OpenAI 的 ChatGPT 模型开发的中文使用版本，专为中文用户，提供更流畅、更精准的智能AI对话。全面体验 ChatGPT 中文版，无需翻墙，支持 GPT-4、GPT-4o、o1、o3 和本地化功能。本项目提供一站式的 ChatGPT中文版使用指南，包括国内可用的 ChatGPT镜像网站，帮助快速上手，无论是个人使用还是专业需求，均可无限使用 GPT-4、GPT-4o、o1、o3及Claude 4 Sonnet、Gemini 2.5 Pro、Grok3~

Sorachio - AI Assistant

My collection of Sora Modules

Habitad for humanity LMS from from zero.

A compassionate, context-aware mental health chatbot built using PDF-based knowledge, vector search (ChromaDB), HuggingFace embeddings, and LLaMA 3-70B via Groq API. The interface is designed with Gradio for a friendly, chat-like user experience.

From zero reproduce mini deepseek-r1 at single gpu 16g

🎉收录精选数字服务资源：虚拟信用卡、AI 工具、云服务器、开发者神器与数字生活必备服务等等... 数字生活,虚拟信用卡,海外充值,chatgpt,ai工具推荐,云服务器,vps推荐,minecraft开服,服务器推荐,免费资源,自托管,免费vps,免费ai,数字工具,雨云,海外手机号,数字游民,推荐合集,优惠推荐,开发者资源

Moxin-VLM: Designed for advanced Vision-Language understanding and interaction, built upon the Moxin-LLM backbone

Stable Diffusion-Enhanced 3D Gaussian Splatting for Efficient Neural Reconstruction of Chinese Heritage Structures

A dataset for studying the semantic differences between synthetic and natural images, generated from shared captions using Stable Diffusion, SDXL, and Infinity models.

Flagship Typhoon model

# text2sql-agentThis project creates an application for querying relational databases using an agent. The agent builds SQL queries, executes them, and returns responses in natural language for a conversational experience. 🐙✨## Folder Structure```plaintext📦 webinar_text2sql├── 📁 chatbot # Text-to-SQL

A GPU-accelerated, physically-based path tracer built with C++ and OpenGL - from scratch.

OpenCompass is an LLM evaluation platform, supporting a wide range of models (Llama3, Mistral, InternLM2,GPT-4,LLaMa2, Qwen,GLM, Claude, etc) over 100+ datasets.

# 🌊 Frequency Separation Extension for WebUIEnhance your Stable Diffusion outputs by processing images in three layers: low, mid, and high frequencies. This method produces sharper and more detailed renders, making your images stand out. 🛠️

# gemma-2-2b-it.csThis project implements int8 CPU inference in pure C#. It ports a Rust repository using Gemini 2.5 Pro Preview, and you can easily build and run it with the provided batch files. 🐙💻

An LLM focused on running locally.

WebIntel is an advanced web intelligence system that delivers real-time insights and analysis. 🚀 It leverages cutting-edge AI technology to enhance your web research experience. 💻

Experimental project... simplest "Deepseek-like" UWP app created with (in) Trae Ai service for my old sweet Windows Family devices! ;)

A comprehensive list of state-of-the-art papers in autonomous driving and robotics, covering end-to-end (e2e) learning, vision-language models (VLM), vision-language-action models (VLA), world models, and reinforcement learning (RL), with links to papers, code repositories, and relevant websites.

An open-source, production-ready AI chat application featuring Claude Sonnet 4, GPT-4o, real-time messaging, and modern UI. Built with Next.js 15, Convex, and TypeScript.

LlamaIndex integration with the ASIMOV Platform.

Eris is a private AI chat application that runs entirely on your device using Apple's MLX framework. Named after the dwarf planet that challenged our understanding of the solar system, Eris challenges the notion that AI must live in the cloud.

MinerU 2.0 VLM 网页应用

A vibe coded Sora JSON Prompt Crafter for curious humans and prompt engineers

Proxy server for Claude Code that converts requests to OpenAI-compatible APIs. Supports multiple providers and offers real-time streaming. 🚀💻

A full-stack web application that uses a chain of AI models (Stable Diffusion, InsightFace, LLMs) to provide a complete workflow for creating complex images, from background removal and face swapping to a full-featured meme editor.

Noto.ai is a local AI-powered PDF assistant that lets you chat with documents, ask questions, and get instant summaries—all offline using LLaMA 3 via Ollama. Built with Python and Kivy, it offers a clean desktop interface for students, researchers, and professionals who want to extract insights from complex PDFs efficiently and privately.