The repo is finally unlocked. enjoy the party! The fastest repo in history to surpass 100K stars ⭐. Join Discord: https://discord.gg/5TUQKqFWd Built in Rust using oh-my-codex.

Code for "Recurrent Independent Mechanisms"

An agentic skills framework & software development methodology that works.

The simplest, fastest repository for training/finetuning medium-sized GPTs.

Build, run and scale AI agents like API and microservices - observable,auditable and identity-aware from day one.

Neuro-Symbolic Agentic Protocol

Generic building-block toolbox for training neural networks with adaptive and recursive execution. It provides reusable components to control iteration, stopping, and unrolling during training, ena…

Our first fully AI generated deep learning system

A flexible framework for AI agents that separates workflow logic from search strategy. Based on https://arxiv.org/pdf/2512.03571 Also read https://asari.ai/blog/encompass

Probabilistic programming with vectorized programmable inference

Probabilistic programming with vectorized programmable inference

🦉 OWL: Optimized Workforce Learning for General Multi-Agent Assistance in Real-World Task Automation

Implementation of the MetaController proposed in "Emergent temporal abstractions in autoregressive models enable hierarchical reinforcement learning" from the Paradigms of Intelligence team at Google

Implementation of GenMimic, "From Generated Human Videos to Physically Plausible Robot Trajectories"

Implementation of Hippoformer, Integrating Hippocampus-inspired Spatial Memory with Transformers

AlphaSuite is an open-source quantitative analysis platform that gives you the power to build, test, and deploy professional-grade trading strategies. It's designed for traders and analysts who wan…

An introduction to theorem proving in Lean for the impatient.

Retrieval and Retrieval-augmented LLMs

Pioneering Automated GUI Interaction with Native Agents

Legible, Scalable, Reproducible Foundation Models with Named Tensors and Jax

We integrate discrete diffusion models with neurosymbolic predictors for scalable and calibrated learning and reasoning

Fine-tune Stable Audio Open with DiT ControlNet.

Modeling Attention and Binding in the Brain through Bidirectional Recurrent Gating

ORB forcefield models from Orbital Materials

Efficiently discovering algorithms via LLMs with evolutionary search and reinforcement learning.

Lets make video diffusion practical!