Stars
Aligning Agentic World Models via Knowledgeable Experience Learning
[NeurIPS2025] "AI-Researcher: Autonomous Scientific Innovation" -- A production-ready version: https://novix.science/chat
InternAgent-1.5: A Unified Agentic Framework for Long-Horizon Autonomous Scientific Discovery
RepoMaster: The open-source AI agent that masters GitHub. It turns any code repository into a powerful tool, achieving a new level of autonomous task-solving. An open alternative to Claude-Code.
MiroFlow is an agent framework that enables tool-use agent tasks, featuring a reproducible GAIA score of 82.4%.
MemGen: Weaving Generative Latent Memory for Self-Evolving Agents
[ICLR 2026] LightMem: Lightweight and Efficient Memory-Augmented Generation
DeepEvolve is a research and coding agent for new algorithm discovery in different science domains with Deep Research and AlphaEvolve.
Democratizing AI scientists with ToolUniverse
[ICLR/AAAI 2026] Open-Source LLM-Based Data Analysis Agents
OceanGym: A Benchmark Environment for Underwater Embodied Agents
gpt-oss-120b and gpt-oss-20b are two open-weight language models by OpenAI
[Lumina Embodied AI] 具身智能技术指南 Embodied-AI-Guide
A powerful tool for creating datasets for LLM fine-tuning 、RAG and Eval
Open AI data scientist agent that automates complex data analysis tasks using the ReAct framework. Execute Python code locally or in the cloud, upload datasets, and generate detailed analytical rep…
KnowRL: Exploring Knowledgeable Reinforcement Learning for Factuality
AutoMind: Adaptive Knowledgeable Agent for Automated Data Science
A Framework for LLM-based Multi-Agent Reinforced Training and Inference
Train your Agent model via our easy and efficient framework
🔧Tool-Star: Empowering LLM-brained Multi-Tool Reasoner via Reinforcement Learning
verl-agent is an extension of veRL, designed for training LLM/VLM agents via RL. verl-agent is also the official code for paper "Group-in-Group Policy Optimization for LLM Agent Training"
[IROS 2025] OceanSim: A GPU-Accelerated Underwater Robot Perception Simulation Framework
[ACL 2025] SynWorld: Virtual Scenario Synthesis for Agentic Action Knowledge Refinement
[EMNLP 2025] LightThinker: Thinking Step-by-Step Compression
Agent-R1: Training Powerful LLM Agents with End-to-End Reinforcement Learning