Stars
A unified framework for easy reinforcement learning in Flow-Matching models
A kernel library written in tilelang
Sim2Reason: Solving Physics Olympiad via Reinforcement Learning on Physics Simulators. We present a method for turning physics simulators into scalable generators of question–answer pairs for impro…
An Asynchronous Reinforcement Learning Engine for Omni-Modal Post-Training at Scale
CUDA kernels for linear attention variants, written in CuTe DSL and CUTLASS C++.
Based on The Minimalist Entrepreneur by Sahil Lavingia
Run OpenClaw more securely inside NVIDIA OpenShell with managed inference
Claw-Eval is an evaluation harness for evaluating LLM as agents. All tasks verified by humans.
[NeurIPS 2025] PhysCtrl: Generative Physics for Controllable and Physics-Grounded Video Generation
Real-Time Physical Action-Conditioned Video Generation
PhysGen: Rigid-Body Physics-Grounded Image-to-Video Generation (ECCV 2024)
GenEval: An object-focused framework for evaluating text-to-image alignment
Open-source evaluation toolkit of large multi-modality models (LMMs), support 220+ LMMs, 80+ benchmarks
Minimalist RL for Diffusion LLMs with SOTA reasoning performance (89.1% GSM8K). Official implementation of "The Flexibility Trap".
This repository contains the official implementation of Physical Parameter-Guided Diffusion (PgD), a framework for generating high-fidelity microbubble point spread functions (PSFs) under diverse u…
slime is an LLM post-training framework for RL Scaling.
Rediscover your social memories with local, AI-powered analysis. 本地化的聊天记录分析工具，通过 AI Agent 回顾你的社交记忆。
Enjoy the magic of Diffusion models!
WeDLM: The fastest diffusion language model with standard causal attention and native KV cache compatibility, delivering real speedups over vLLM-optimized baselines.
SANA: Efficient High-Resolution Image Synthesis with Linear Diffusion Transformer
A framework for efficient model inference with omni-modality models
The repository provides code for running inference and finetuning with the Meta Segment Anything Model 3 (SAM 3), links for downloading the trained model checkpoints, and example notebooks that sho…
PyTorch implementation of JiT https://arxiv.org/abs/2511.13720
Enable macOS HiDPI and have a native setting.
[NeurIPS 2025] Improving Video Generation with Human Feedback