Discover The Top AI hackers & Projects

Get started with building Fullstack Agents using Gemini 2.5 and LangGraph

Collection of extracted System Prompts from popular chatbots like ChatGPT, Claude & Gemini

The simplest, fastest repository for training/finetuning small-sized VLMs.

PandaWiki 是一款 AI 大模型驱动的开源知识库搭建系统，帮助你快速构建智能化的 产品文档、技术文档、FAQ、博客系统，借助大模型的力量为你提供 AI 创作、AI 问答、AI 搜索等能力。

A Model Context Protocol (MCP) Gateway & Registry. Serves as a central management point for tools, resources, and prompts that can be accessed by MCP-compatible LLM applications. Converts REST API endpoints to MCP, composes virtual MCP servers with added security and observability, and converts between protocols (stdio, SSE, Streamable HTTP).

Open-source platform for AI-driven storytelling, worldbuilding, and immersive roleplay

The world's first Full-Stack Open-Source General AI Agent

AI 洞察日报项目，每日为您精选 AI 领域的最新动态，包括行业新闻、热门开源项目和前沿学术论文以及科技大V推文，并通过 Google Gemini 模型进行智能处理与日报生成，最终自动发布到 GitHub Pages 生成AI日报。

VLM-3R: Vision-Language Models Augmented with Instruction-Aligned 3D Reconstruction

An example agent demonstrating streaming, tool use, and interactivity from your terminal. This agent builder can help you to build your own agents and tools.

Pixel-Level Reasoning Model trained with RL

Adapting VLMs to Bench2Drive.

AetherLink移动应用是一个基于现代Web技术构建的跨平台AI助手应用。该应用支持与多种AI模型（如OpenAI、Google Gemini、Anthropic Claude、Grok、硅基流动、火山方舟等）的交互，提供流畅的对话体验，并支持Android平台部署。应用采用React、TypeScript和Capacitor框架开发，具有高度可定制的模型配置、多主题聊天管理、AI思考过程可视化、语音合成、语音识别、MCP工具支持、知识库管理等特色功能。

Local LLM Server with GPU and NPU Acceleration

CHATS: Combining Human-Aligned Optimization and Test-Time Sampling for Text-to-Image Generation (ICML2025)

Code for the paper "VTool-R1: VLMs Learn to Think with Images via Reinforcement Learning on Multimodal Tool Use"

Train Large Language Models on MLX.

基于 Google Gemini AI 的全功能图像处理应用

Easiest way to give context to LLMs; Attachments has the ambition to be the general funnel for any files to be transformed into images+text for large language models context by only adding 2 lines to your python code.

Repository for awesome spatial/visual reasoning MLLMs. (focus more on embodied applications)

Github repository for "Bring Reason to Vision: Understanding Perception and Reasoning through Model Merging" (ICML 2025)

Vision Language Models are Biased

Code2Logic: Game-Code-Driven Data Synthesis for Enhancing VLMs General Reasoning

An implementation of GRPO for Unsloth's VLMs training

Take your LLM to the optometrist.

A comprehensive list of awesome research, resources, and tools for leveraging LLMs, VLMs, and VLA models in autonomous driving decision-making and motion planning.

Llama.cpp runner/swapper and proxy that emulates LMStudio / Ollama backends

cadrille: Multi-modal CAD Reconstruction with Online Reinforcement Learning

This is code repo for the paper, Depolying Ten Thousand Robots: Scalable Imitation Learninig for Lifelong Mulit-Agent Path Finding, which won the ICRA 2025 best paper on multi-robot systems and the best student paper.

system prompts of LLMs, AI Tools, AI Products, DeepSeek,ChatGPT, Gemini, Grok, Qwen

Concept for a node-based memory system for LLMs

A Recipe for Building LLM Reasoners to Solve Complex Instructions

a comprehensive diffusion-based code refinement model

A Commitizen plugin that leverages OpenAI's GPT-4o to automatically generate clear, concise, and conventional commit messages based on your staged git changes.

A modern, fast, and beautiful Emacs configuration for developers who want power without the complexity.

Synthetic Dialog Generation and Analysis with LLMs

🧠 Train your own DeepSeek-R1 style reasoning model on Mac! First MLX implementation of GRPO - the breakthrough technique behind R1's o1-matching performance. Build mathematical reasoning AI without expensive RLHF. Apple Silicon optimized. 🚀

ComfyUI-AniSora is now available in ComfyUI, Index-AniSora is the most powerful open-source animated video generation model. It enables one-click creation of video shots across diverse anime styles including series episodes, Chinese original animations, manga adaptations, VTuber content, anime PVs, mad-style parodies(鬼畜动画), and more!

TIME: A Multi-level Benchmark for Temporal Reasoning of LLMs in Real-World Scenario

ComfyUI custom node powered by LLaVA VLM for advanced image captioning with customizable styles and memory-efficient inference.

A simplified mlx implementation of the original stable diffusion 1.5 algorithm featuring fine-tuned weight loading and lora loading for Text and Image generation