Aman Salykov

aiter. AI Tensor Engine for ROCm

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

HipKittens. Fast and Furious AMD Kernels

hf-rocm-kernels. Python

Direct3D-S2. [NeurIPS 2025] Direct3D‑S2: Gigascale 3D Generation Made Easy with Spatial Sparse Attention

flash-attention. Fast and memory-efficient exact attention

modular. The Modular Platform (includes MAX & Mojo)

OmniGen2. OmniGen2: Exploration to Advanced Multimodal Generation. https://arxiv.org/abs/2506.18871

nano-vllm. Nano vLLM

llm.c. LLM training in simple, raw C/CUDA

LMCache. LMCache: Supercharge Your LLM with the Fastest KV Cache Layer

visionbook. <Foundations of Computer Vision> Book

fastgen. Simple high-throughput inference library

octgpt. OctGPT: Octree-based Multiscale Autoregressive Models for 3D Shape Generation [SIGGRAPH 2025]

DDT. [CVPR 2026] DDT: Decoupled Diffusion Transformer

svraster. [CVPR 2025] Sparse Voxels Rasterization: Real-time High-fidelity Radiance Field Rendering

vggt. [CVPR 2025 Best Paper Award] VGGT: Visual Geometry Grounded Transformer

DyT. Code release for DynamicTanh (DyT)

imm. Official implementation of Inductive Moment Matching

cutlass. CUDA Templates and Python DSLs for High-Performance Linear Algebra

Direct3D. [NeurIPS 2024] Direct3D: Scalable Image-to-3D Generation via 3D Latent Diffusion Transformer

Data-Structures-and-Algorithms. A repository that contains all the Data Structures and Algorithms concepts and their implementation in several ways, programming questions and Interview questions. The main aim of this repository is to help students who are learning Data Structures and Algorithms or preparing for an interview.

CUT3R. Official implementation of Continuous 3D Perception Model with Persistent State

radfoam. Original implementation of "Radiant Foam: Real-Time Differentiable Ray Tracing"

SCube. [NeurIPS 2024] SCube: Instant Large-Scale Scene Reconstruction using VoxSplats

cuda-tensorcore-hgemm. Cuda

micro_diffusion. Official repository for our work on micro-budget training of large-scale diffusion models.

genesis-world. Simulation platform for general-purpose robotics & embodied AI learning.

StableAnimator. [CVPR2025] We present StableAnimator, the first end-to-end ID-preserving video diffusion framework, which synthesizes high-quality videos without any post-processing, conditioned on a reference image and a sequence of poses.

PSHuman. Python

TRELLIS. Official repo for paper "Structured 3D Latents for Scalable and Versatile 3D Generation" (CVPR'25 Spotlight).

smol-course. A course on aligning smol models.

GaussianSpeech. [ICCV 2025] GaussianSpeech: Audio-Driven Gaussian Avatars

GaussianAvatars. [CVPR 2024 Highlight] The official repo for "GaussianAvatars: Photorealistic Head Avatars with Rigged 3D Gaussians"

NPHM. [CVPR'23] Learning Neural Parametric Head Models

MultiPly. MultiPly: Reconstruction of Multiple People from Monocular Video in the Wild (CVPR2024 Oral)

SPARK. Official implementation for the SIGGRAPH Asia 2024 paper SPARK: Self-supervised Personalized Real-time Monocular Face Capture

Sana. SANA: Efficient High-Resolution Image Synthesis with Linear Diffusion Transformer

EdgeRunner. [ICLR 2025] EdgeRunner: Auto-regressive Auto-encoder for Efficient Mesh Generation

SurFhead. .

lingua. Meta Lingua: a lean, efficient, and easy-to-hack codebase to research LLMs.

drrobot. Code for "Differentiable Robot Rendering" (CoRL 2024)

hart. HART: Efficient Visual Generation with Hybrid Autoregressive Transformer

GAGAvatar. [NeurIPS 2024] Generalizable and Animatable Gaussian Head Avatar

GS-VTON. [IJCV 2026] GS-VTON: Controllable 3D Virtual Try-on with Gaussian Splatting

uv. An extremely fast Python package and project manager, written in Rust.

cuda_hgemm. Several optimization methods of half-precision general matrix multiplication (HGEMM) using tensor core with WMMA API and MMA PTX instruction.

cuda_hgemv. Several optimization methods of half-precision general matrix vector multiplication (HGEMV) using CUDA core.

modern-cpp-features. A cheatsheet of modern C++ language and library features.

mar. PyTorch implementation of MAR+DiffLoss https://arxiv.org/abs/2406.11838

YHs_Sample. Yinghan's Code Sample

how-to-optimize-gemm. row-major matmul optimization

FP16. Conversion to/from half-precision floating point formats

hardware-effects-gpu. Demonstration of various hardware effects on CUDA GPUs.

xlstm. Official repository of the xLSTM.

dust3r. DUSt3R: Geometric 3D Vision Made Easy

LoG. Level of Gaussians

inference-optimization-blog-post. Jupyter Notebook

2d-gaussian-splatting. [SIGGRAPH'24] 2D Gaussian Splatting for Geometrically Accurate Radiance Fields

resource-stream. GPU programming related news and material links

lectures. Material for gpu-mode lectures

gaussian_surfels. [SIGGRAPH'24] Implementations for "High-quality Surface Reconstruction using Gaussian Surfels".

torchtitan. A PyTorch native platform for training generative AI models

tinygrad. You like pytorch? You like micrograd? You love tinygrad! ❤️

llama3. The official Meta Llama 3 GitHub site

coding-interview-university. A complete computer science study plan to become a software engineer.

warp. A Python framework for GPU-accelerated simulation, robotics, and machine learning.

awesome-virtual-try-on. A curated list of awesome research papers, projects, code, dataset, workshops etc. related to virtual try-on.

Transformers-Tutorials. This repository contains demos I made with the Transformers library by HuggingFace.

dolma. Data and tools for generating and inspecting OLMo pre-training data.

executorch. On-device AI across mobile, embedded and edge for PyTorch

aim. Aim 💫 — An easy-to-use & supercharged open-source experiment tracker.

3dv_tutorial. An Invitation to 3D Vision: A Tutorial for Everyone

float8_experimental. This repository contains the experimental PyTorch native float8 training UX

Pearl. A Production-ready Reinforcement Learning AI Agent Library brought by the Applied Reinforcement Learning team at Meta.

FoundationPose. [CVPR 2024 Highlight] FoundationPose: Unified 6D Pose Estimation and Tracking of Novel Objects

AlpacaDataCleaned. Alpaca dataset from Stanford, cleaned and curated

litgpt. 20+ high-performance LLMs with recipes to pretrain, finetune and deploy at scale.

gpt-fast. Simple and efficient pytorch-native transformer text generation in <1000 LOC of python.

CtCI-6th-Edition. Cracking the Coding Interview 6th Ed. Solutions

xformers. Hackable and optimized Transformers building blocks, supporting a composable construction.

llama-cookbook. Welcome to the Llama Cookbook! This is your go to guide for Building with Llama: Getting started with Inference, Fine-Tuning, RAG. We also show you how to solve end to end problems using Llama model family and using them on various provider services

llama. Inference code for Llama models

rl-baselines3-zoo. A training framework for Stable Baselines3 reinforcement learning agents, with hyperparameter optimization and pre-trained agents included.

Wallpapers. My current wallpaper rotation

pycolmap. (Deprecated) Python bindings for COLMAP

ruff. An extremely fast Python linter and code formatter, written in Rust.

Open3D. Open3D: A Modern Library for 3D Data Processing

annotated_deep_learning_paper_implementations. 🧑‍🏫 60+ Implementations/tutorials of deep learning papers with side-by-side notes 📝; including transformers (original, xl, switch, feedback, vit, ...), optimizers (adam, adabelief, sophia, ...), gans(cyclegan, stylegan2, ...), 🎮 reinforcement learning (ppo, dqn), capsnet, distillation, ... 🧠

AITemplate. AITemplate is a Python framework which renders neural network into high performance CUDA/HIP C++ code. Specialized for FP16 TensorCore (NVIDIA GPU) and MatrixCore (AMD GPU) inference.

scalene. Scalene: a high-performance, high-precision CPU, GPU, and memory profiler for Python with AI-powered optimization proposals

cpp-cheatsheet. Modern C++ Cheatsheet

openMVG. open Multiple View Geometry library. Basis for 3D computer vision and Structure from Motion.

openMVS. open Multi-View Stereo reconstruction library

example-code-2e. Example code for Fluent Python, 2nd edition (O'Reilly 2022)

awesome-internships. A curated list of tech internships resources.

awesome-interview-questions. :octocat: A curated awesome list of lists of interview questions. Feel free to contribute! :mortar_board:

CtCI-6th-Edition-cpp. Cracking the Coding Interview 6th Ed. C++ Solutions

CtCI-6th-Edition-Python. Cracking the Coding Interview 6th Ed. Python Solutions