A technical exploration comparing standard deep RL (PPO) against biologically plausible learning rules on a custom Pong environment. Everything is implemented from scratch with no Stable-Baselines,…

HRM-Text is a 1B text generation model based on the HRM architecture, strengthened by task completion and latent space reasoning.

Isaac Lab API, powered by MuJoCo-Warp, for RL and robotics research

A platform for reproducible world model research and evaluation

Explorations into the proposal from the paper, Visual-tactile pretraining and online multitask learning for humanlike manipulation dexterity

Implementing scalable LLMs in pure JAX (no third-party libraries)

[AES'23] A Fast Modular Synthesizer in JAX

Rust implementation of Tiny Recursive Models for efficient puzzle solving

Goal is to solve sample efficiency by using ARC-AGI as a benchmark

NVARC solution to ARC-AGI-2

Continuous Thought Machines, because thought takes time and reasoning is a process.

A Graph Neural Network Library in Jax

Code for PuzzleJAX, a benchmark for reasoning and learning, that reimplements PuzzleScript, a concise and expressive DSL and game engine for grid-based puzzle games, in JAX.

An MDL-based approach to the Abstraction and Reasoning Corpus (ARC) challenge

Abstract Reasoning with Graph Abstractions (ARGA) implementation

Official Project Page for HLA: Higher-order Linear Attention (https://arxiv.org/abs/2510.27258)

Thermodynamic Hypergraphical Model Library in JAX

HRM Agent repo

Unofficial implementation of Tiny Recursive Model (TRM), improvement to HRM from Sapient AI, by Alexia Jolicoeur-Martineau

A comprehensive JAX/NNX library for diffusion and flow matching generative algorithms, featuring DiT (Diffusion Transformer) and its variants as the primary backbone with support for ImageNet train…

Message passing for expected free energy minimizing agent based on multivariate autoregressive exogenous models.