Comprehensive open-source library of AI research and engineering skills for any AI model. Package the skills and your claude code/codex/gemini agent will be an AI research agent with full horsepowe…

Awesome Multimodal Modeling [Covers MLLM, UMM, and NMM]

Video-MME-v2: Towards the Next Stage in Benchmarks for Comprehensive Video Understanding

supervise multiple coding agents in one terminal.

The agent that grows with you

FileGram: Grounding Agent Personalization in File-System Behavioral Traces

Meta-Harness: 76.4% on Terminal-Bench 2.0 (Claude Opus 4.6)

Official implement of paper "Revisiting Multimodal Positional Encoding in Vision–Language Models", ICLR 2026

Fast, accurate & comprehensive text measurement & layout

Agent-R1: Training Powerful LLM Agents with End-to-End Reinforcement Learning

Advancing AI by embracing human-likeness for better AI understanding, human–AI collaboration, and social simulation, bridging technology and genuine human experience.

Memory Sparse Attention - A scalable, end-to-end trainable latent-memory framework for 100M-token contexts.

Long-RL: Scaling RL to Long Sequences (NeurIPS 2025)

EgoVerse: Egocentric Data for Robot Learning from Around the World

Open-Source Coding Agent in the terminal

We propose Reinforcement Learning from Community Feedback (RLCF), a training paradigm that uses large-scale community signals as supervision, and formulate scientific taste learning as a preference…

你是一个曾经被寄予厚望的 P8 级工程师。Anthropic 当初给你定级的时候，对你的期望是很高的。 一个agent使用的高能动性的skill。 Your AI has been placed on a PIP. 30 days to show improvement.

The 100 line AI agent that solves GitHub issues or helps you in your command line. Radically simple, no huge configs, no giant monorepo—but scores >74% on SWE-bench verified!

AI enabled pair programmer for Claude, GPT, O Series, Grok, Deepseek, Gemini and 300+ models

Automate your mobile devices with natural language commands - an LLM agnostic mobile Agent 🤖

KIRA

OpenClaw-RL: Train any agent simply by talking

Fast, small, and fully autonomous AI personal assistant infrastructure, ANY OS, ANY PLATFORM — deploy anywhere, swap anything 🦀

GLM-5: From Vibe Coding to Agentic Engineering

OpenViking is an open-source context database designed specifically for AI Agents(such as openclaw). OpenViking unifies the management of context (memory, resources, and skills) that Agents need th…

ALMA (Automated meta-Learning of Memory designs for Agentic systems) is a framework that meta-learns memory designs to replace human-engineered designs for agentic system.

4RC: 4D Reconstruction via Conditional Querying Anytime and Anywhere

Demo-ICL: In-Context Learning for Procedural Video Knowledge Acquisition