EmbraceAGI/Awesome-AGI

A curated list of awesome AGI frameworks, software and resources

What it solves

The AGI field is moving fast, with new models, frameworks, and papers appearing constantly. This project solves the problem of scattered information by gathering the most important AGI resources into one organized list. Instead of hunting across many sites, you can find frameworks, agent harnesses, frontier models, papers, and demos here.

How it works

It's an "awesome list" — a single Markdown document with categorized sections. The list is maintained and periodically updated (the README notes a refresh in 2026). Each entry typically includes the project name, GitHub stars, and a short description. Sections cover definitions of AGI, frontier models, agent harnesses, frameworks, agents, papers, datasets, blogs, and websites. There are also links to online demos and companion lists.

Who it’s for

This is for developers, researchers, and AI enthusiasts who want to explore the AGI ecosystem. Whether you're looking for a coding agent, a multi-agent framework, a frontier model to try, or academic papers, this list gives you a starting point. It's also useful for keeping an eye on the latest developments.

Highlights

  • Includes major autonomous agent frameworks like AutoGPT, MetaGPT, and AgentGPT.
  • Curates frontier models from 2026, such as Claude 5, GPT-5.6, Gemini 3.7, DeepSeek-V4, and more.
  • Features agent harnesses like Claude Code, OpenAI Codex, and deepseek-harness.
  • Lists research papers like "Generative Agents" and "AgentBench".
  • Provides links to online demos (Cognosys, Godmode) and additional resources like LangGPT.

Related

  • Project
    blakeandersonw/ai-app-playbook

    A comprehensive guide and technology stack for building, launching, and scaling profitable AI applications while minimizing infrastructure costs.

  • Project
    MaxFreedomPollard/Compartment

    An encrypted, fully offline persistent memory vault for AI agents that allows them to share learned facts and preferences across sessions without any network access or telemetry.

  • Project
    OpenMinis/MinisSkills

    MinisSkills is a community collection of reusable skills for Claude, each packaged as a SKILL.md file that extends the assistant with specialized workflows, domain knowledge, and optional bundled resources like scripts and references. It solves the problem of giving Claude domain-specific expertise on demand through a standardized skill format.

  • Project
    huggingface/funes

    Funes is a durable memory system for AI coding agents that indexes past sessions and decisions into a searchable dataset, allowing agents to recall project history and shareable knowledge via the Hugging Face Hub.