Starred repositories
A collection of 100+ specialized Claude Code subagents covering a wide range of development use cases
A curated list for Self-Improvement in Foundation Model Based Agentic Systems.
NVIDIA Cosmos is an open platform of world models, datasets, and tools that enables developers to build Physical AI for robots, autonomous vehicles, smart infrastructure, and more.
AI agents running research on single-GPU nanochat training automatically
Fully autonomous & self-evolving research from idea to paper. Chat an Idea. Get a Paper. 🦞
Open-source implementation of AlphaEvolve
A Survey of Reinforcement Learning for Large Reasoning Models
Fine-tune LLM agents with online reinforcement learning
NumPy+Jax with named axes and an uncompromising attitude
An open-source alternative to OpenAI and Gemini's deep research.
A template for writing academic papers in Markdown.
A high-throughput and memory-efficient inference and serving engine for LLMs
Elegant easy-to-use neural networks + scientific computing in JAX. https://docs.kidger.site/equinox/
Online Goal-Conditioned Reinforcement Learning in JAX. ICLR 2025 Spotlight.
🏛️A research-friendly codebase for fast experimentation of single-agent reinforcement learning in JAX • End-to-End JAX RL
Really Fast End-to-End Jax RL Implementations
PDF references add-on for Zotero.
Implements QuickLook in Zotero
Scripts to build a trimmed-down Windows 11 image.
Safe Reinforcement Learning with Natural Language Constraints
High-quality single file implementation of Deep Reinforcement Learning algorithms with research-friendly features (PPO, DQN, C51, DDPG, TD3, SAC, PPG)
PyTorch implementation of Advantage Actor-Critic (A2C), Asynchronous Advantage Option-Critic (A2OC), Proximal Policy Optimization (PPO) and Scalable trust-region method for deep reinforcement learn…