Collections of robotics environments geared towards benchmarking multi-task and meta reinforcement learning

PyTorch implementation of JiT https://arxiv.org/abs/2511.13720

A Collection of Competitive Text-Based Games for Language Model Evaluation and Reinforcement Learning

Open-source evaluation toolkit of large multi-modality models (LMMs), support 220+ LMMs, 80+ benchmarks

RL gym for vision language models written in JAX

This repo contains the source code for the paper "Evolution Strategies at Scale: LLM Fine-Tuning Beyond Reinforcement Learning"

MobileLLM-R1

Experiment for abstraction learning

Train your Agent model via our easy and efficient framework

Open-source framework for the research and development of foundation models.

Hierarchical Reasoning Model Official Release

KernelBench: Can LLMs Write GPU Kernels? - Benchmark + Toolkit with Torch -> CUDA (+ more DSLs)

SkyReels-V2: Infinite-length Film Generative model

Interactive visualizations of the geometric intuition behind diffusion models.

User-friendly implementation of the Mixture-of-Sparse-Attention (MoSA). MoSA selects distinct tokens for each head with expert choice routing providing a content-based sparse attention mechanism.

Making large AI models cheaper, faster and more accessible

Lets make video diffusion practical!

Qwen3 is the large language model series developed by Qwen team, Alibaba Cloud.

[ICML2025] SpargeAttention: A training-free sparse attention that accelerates any model inference.

Towards Human-Sounding Speech

Official PyTorch implementation for Hogwild! Inference: Parallel LLM Generation with a Concurrent Attention Cache

Official PyTorch implementation of One-Minute Video Generation with Test-Time Training

UniDisc: A discrete diffusion model for joint multimodal generation, enabling controllable and efficient text-image synthesis, editing, and inpainting.

Reproduction of Dreamerv1 and v2 in pytorch for deepmind control suite

Agent S: an open agentic framework that uses computers like a human