AI & ML · Daily GitHub chart, Tue, Nov 17

AI & ML repositories pooled from every language chart and re-ranked by stars gained. As of Tue, Nov 17, 9 PM GMT+9.

  1. 1

    google-research/paper-orchestra

    Multi-agent framework for automated AI research paper writing

    PaperOrchestra is a multi-agent framework that turns raw research materials such as ideas and experimental logs into submission-ready LaTeX manuscripts. Specialized agents handle outline generation, literature review, section writing, content refinement, and plotting.

    Python4.7k starsAI & ML

    +313that day

  2. 2

    elie222/rakazo

    Open-source self-hosted AI bot platform

    Rakazo is an open-source, self-hosted alternative to Grok Bot where you choose your own AI models and sandboxes. Each bot gets persistent memory, routines, and a computer with a browser, terminal, and files.

    TypeScript9.2k starsAI & ML

    +266that day

  3. 3

    jingyaogong/minimind

    64M-parameter LLM trained from scratch in 2 hours

    MiniMind is an open-source project for training a 64M-parameter language model from scratch using native PyTorch. It implements the full LLM pipeline—from data cleaning and pretraining to SFT, reinforcement learning, and tool use—without relying on high-level wrapper libraries.

    Python66.6k stars1 good first issueAI & ML

    +188that day

  4. 4

    Helldez/BigMoeOnEdge

    Streaming inference engine for oversized MoE models

    A llama.cpp-based inference engine that runs Mixture-of-Experts models larger than device memory by streaming needed experts from flash storage. It includes desktop and mobile apps alongside an OpenAI- and Anthropic-compatible local server.

    C++1k starsAI & ML

    +98that day

  5. 5

    a-ghorbani/pocketpal-ai

    On-device AI assistant app for phones

    A React Native mobile app that runs language models directly on your phone without an account or internet connection. It handles chat, voice input and output, and tool use locally on the device.

    TypeScript9.6k starsAI & ML

    +85that day

  6. 6

    QuantumNous/new-api

    Self-hosted open-source AI gateway

    A Go-based self-hosted AI gateway that connects upstream model services and exposes unified OpenAI-, Anthropic-, and Gemini-compatible APIs. It centralizes model routing, access control, usage tracking, and cost accounting with a built-in web console.

    Go52.2k stars1 good first issueAI & ML

    +71that day

  7. 7

    vstorm-co/agenticos

    Open-source self-hosted AI agent platform for teams

    AgenticOS is an open-source, self-hosted platform built on Pydantic AI for building, running, and governing AI teammates on your own infrastructure. It brings together shared chat, RAG, skills, a real browser, and 4,700+ MCP servers in one place.

    Python417 starsAI & ML

    +59that day

  8. 8

    NVIDIA-NeMo/Switchyard

    LLM model routing library for AI agents

    Switchyard is an open-source Rust library that helps AI agents choose which model handles each request. It combines efficient and more capable models to balance task success, cost, and latency.

    Rust3.7k starsAI & ML

    +41that day

  9. 9

    ArvinLovegood/go-stock

    Local-first AI stock analysis desktop app

    Built with Go, Wails, and NaiveUI, this desktop tool combines LLMs with stock market analysis and screening. It stores data locally and supports A-shares, Hong Kong stocks, US stocks, and ETFs.

    Go8.2k starsAI & ML

    +29that day

  10. 10

    mastra-ai/mastra

    TypeScript framework for AI applications and agents

    Mastra is a TypeScript framework for building AI-powered applications and autonomous agents. It covers model routing, workflows, context management, and production evaluation in one stack.

    TypeScript29.6k starsAI & ML

    +25that day

  11. 11

    google/ax

    Declarative orchestration runtime for autonomous agent workloads

    Agent Executor (AX) is Google's open agentic orchestration runtime built on Agent Substrate to run autonomous agent workloads in sandboxed environments at cluster scale. It uses Kubernetes-style declarative YAML manifests (Task, Workspace, and Model) to wire up workspaces and manage agent lifecycles.

    Go14.8k starsAI & ML

    +22that day

  12. 12

    basketikun/chatgpt2api

    Go-based OpenAI-compatible API proxy for ChatGPT web

    A Go-based proxy that reverse-engineers the ChatGPT web interface into OpenAI-compatible API endpoints. It supports image generation and editing, text responses, speech synthesis and transcription, along with a built-in web workspace and account pool management.

    Go6.9k starsAI & ML

    +18that day

  13. 13

    drumih/turbo-fieldfare

    On-device Gemma 4 26B-A4B inference runtime for 8 GB Macs

    TurboFieldfare is a Swift and Metal inference runtime that runs the Gemma 4 26B-A4B model in about 2 GB of RAM on Apple Silicon Macs. It keeps the shared core and KV cache in memory while streaming only the needed experts per token from SSD.

    Swift7.4k starsAI & ML

    +14that day

  14. 14

    dograh-hq/dograh

    Open-source self-hosted voice AI agent platform

    Dograh is an open-source, self-hostable voice AI platform that serves as an alternative to Vapi and Retell. It lets you build production voice agents with a visual workflow builder and connect your own LLM, TTS, STT, and telephony providers.

    Python6.2k starsAI & ML

    +13that day

  15. 15

    HKUDS/RAG-Anything

    All-in-one multimodal RAG framework built on LightRAG

    RAG-Anything is a multimodal RAG framework that parses and queries documents containing text, images, tables, and mathematical equations in a single pipeline. Built on LightRAG, it handles document parsing, knowledge graph construction, and hybrid retrieval through one interface.

    Python24.6k starsAI & ML

    +9that day

  16. 16

    microsoft/semantic-kernel

    Model-agnostic SDK for AI agents and multi-agent systems

    Semantic Kernel is a model-agnostic SDK for building, orchestrating, and deploying AI agents and multi-agent systems. It supports Python, .NET, and Java, with Microsoft Agent Framework serving as its successor.

    C#28.8k starsAI & ML

    +8that day

  17. 17

    ddalcu/mlx-serve

    Native local AI inference server and app for Apple Silicon

    A native AI inference server and macOS app built with a Zig backend and Swift frontend that runs without Python on Apple Silicon. It supports chat, coding agents, images, video, music, speech, and 3D models offline while speaking OpenAI, Anthropic, and Ollama APIs.

    Zig2.5k starsAI & ML

    +8that day