Stars
Official implementation of "Continuous Autoregressive Language Models"
Distributed fine-tuning pipeline for LLaMA 3.2 1B across 8 GPUs using PyTorch DDP
Unofficial implementation and extension of GfR (RSS 2026) and HIL (TOG 2026)
General technology for enabling AI capabilities w/ LLMs and MLLMs
Official code for HiLS-Attention
Code release for "i1: A Simple and Fully Open Recipe for Strong Text-to-Image Models"
Why is LLM inference slow — and how do you make it fast? A hands-on, first-principles course: roofline → KV cache → quantization → parallelism → vLLM/SGLang, with GPU labs on open models.
Per-patch loss-coupling via Soft-MoE dispatch weights for multi-objective masked image modeling (ECCV 2026)
Solve puzzles. Improve your pytorch.
The official implementation of "DynamicVLA: A Vision-Language-Action Model for Dynamic Object Manipulation". (arXiv 2601.22153)
Robotic Mapping and Localization Course
Open source VLA Model powered by NVIDIA Foundation Models
Schedule-Free Optimization in PyTorch
Real-time stream editing pipeline powered by the FLUX.2-klein-4B model, optimized for consumer GPUs
A curated collection of papers, technical reports, frameworks, and tools for on-policy distillation (OPD) of large language models
[ICML 2017] TensorFlow code for Curiosity-driven Exploration for Deep Reinforcement Learning
Source code for SWIFT, an efficient reward model.
A sequence-to-sequence voice conversion toolkit.
TIPSv2 (CVPR'26) and TIPS (ICLR'25)
PyTorch implementation of consistency regularization methods for semi-supervised learning
Why Do We Need Weight Decay in Modern Deep Learning? [NeurIPS 2024]
Lean formalizations for the paper "ABC implies that Ramanujan's Tau function misses almost all primes"