Lists (26)
Books
1 repository
CUDA
7 repositories
DND
50 repositories
Edge
2 repositories
Edge AI
2 repositories
Elixir
1 repository
Index / Survey / Whatever
12 repositories
Inference
7 repositories
Inference Engines
1 repository
JAX
10 repositories
Malachite
1 repository
ML Compilers
4 repositories
MLSys
22 repositories
Netherdeep
82 repositories
Network Science
2 repositories
Performance Optimization
3 repositories
Profiling
21 repositories
Quantization
2 repositories
Recommender Systems
3 repositories
Reinforcement Learning
47 repositories
Research
55 repositories
Stuff
529 repositories
Tools
6 repositories
Triton
35 repositories
Video Diffusion
13 repositories
Visualization
1 repository
Starred repositories
A minimal GPU design in Verilog to learn how GPUs work from the ground up
Fast and accurate automatic speech recognition (ASR) for edge devices
Scaling Interactive World Models to 1000-Frame Horizons via Pose-Free Hierarchical Memory
Repository hosting code to reproduce our paper (with Stanford and TogetherAI), "Making Databases Faster with LLM Evolutionary Sampling"
FlashTile is a CUDA Tile IR compiler that is compatible with NVIDIA's tileiras, targeting SM70 through SM121 NVIDIA GPUs.
What LLMs Think When You Don’t Tell Them What to Think About?
This repository contains companion software for the Colfax Research paper "Categorical Foundations for CuTe Layouts".
Collection of high-quality robo learning papers for bipedal robots.
GPU matmul kernel exploration: Triton, cuTile, and TileIR backend benchmarks on Blackwell
TRUST is a thermohydraulic software package for CFD simulations. It was originally designed for incompressible single-phase and Low Mach Number flows, but now also allows simulating real compressib…
A modern, C++-native, test framework for unit-tests, TDD and BDD - using C++14, C++17 and later (C++11 support is in v2.x branch, and C++03 on the Catch1.x branch)
🔥 Clone and recreate any website as a modern React app in seconds
Official Repository of VisGym: Diverse, Customizable, Scalable Environments for Multimodal Agents
🌎 Regional-to-global coupled ocean and sea ice simulations based on Oceananigans
Google TPU optimizations for transformers models
UCCL is an efficient communication library for GPUs, covering collectives, P2P (e.g., KV cache transfer, RL weight transfer), and EP (e.g., GPU-driven)
FlashInfer Bench @ MLSys 2026: Building AI agents to write high performance GPU kernels
Review automated kernel generation in the era of LLMs
Our first fully AI generated deep learning system
RLinf: Reinforcement Learning Infrastructure for Embodied and Agentic AI
Algorithm powering the For You feed on X