Starred repositories
Train transformer language models with reinforcement learning.
Open-source repo for applying continual learning to autoresearch with SDPO and other RL algorithms. Current example: Qwen3-14B achieves 1.023 val_bpb (−3.1%), surpassing the original Karpathy agent…
Official Codebase for "Neural Thickets: Diverse Task Experts Are Dense Around Pretrained Weights"
Train the smallest LM you can that fits in 16MB. Best model wins!
Code for "Matching Features, Not Tokens: Energy-Based Fine-Tuning of Language Models".
🏝️ OASIS: Open Agent Social Interaction Simulations with One Million Agents.
Digital Red Queen: Adversarial Program Evolution in Core War with LLMs
A repository for major/influential FEP and active inference papers.
A wrapper around the Craftax agent benchmark, for evaluating digital agents over extremely long time horizons
Ghostty-based macOS terminal with vertical tabs and notifications for AI coding agents
[NeurIPS '25] Knowledge Graph Generation from Any Text
Turn any collection of documents into a knowledge graph. Extract entities and relationships via LLM, deduplicate with your approval. Map domains, find hidden connections, spot patterns across docum…
React components for visualizing traces from AI agents
Stanford NLP Python library for benchmarking the utility of LLM interpretability methods
Code Editor for the AI Agents Era - Run an army of Claude Code, Codex, etc. on your machine
Implementation of Prompt-Singer: Controllable Singing-Voice-Synthesis with Natural Language Prompt (NAACL'24).
verl: Volcano Engine Reinforcement Learning for LLMs
An Autonomous Curriculum Reinforcement Learning framework that steers agents to continually learn in specific environments with zero human data.
SkyRL: A Modular Full-stack RL Library for LLMs
OpenClaw-RL: Train any agent simply by talking
RLAnything & DemyAgent: General and scalable agentic RL algorithms across terminal, GUI, SWE, and tool-call settings