Marcel Rød Claim this profile

Language model tokenization at GB/s

No description.

No description.

Implementation of K nearest neighbor using K-D trees with support for CPU and TPU

No description.

No description.

No description.

No description.

Practical assignments for CS349H - Software Techniques for Emerging Hardware Platforms

No description.

No description.

No description.

No description.

Python implementation of Rust's dbg! macro using CSTs

Implementation of RGCNs in PyTorch

No description.

No description.

A toolkit to optimize ML models for deployment for Keras and TensorFlow, including quantization and pruning.

Fetches sample input and output and compares program output.

A small program for fetching and classifying previous exams.

Learn GPU Programming in Mojo🔥 by Solving Puzzles

Graph Neural PDEs

A native, user-mode, multi-process, graphical debugger.

No description.

Universal Tensor Operations in Einstein-Inspired Notation for Python.

🌊 The ultimate Fish prompt.

No description.

[ICLR 2026] RF-DETR is a real-time object detection and segmentation model architecture developed by Roboflow, SOTA on COCO, designed for fine-tuning.

Accelerate, Optimize performance with streamlined training and serving options with JAX.

Dataframes powered by a multithreaded, vectorized query engine, written in Rust

Unsupervised text tokenizer for Neural Network-based text generation.

TensorRT-LLM provides users with an easy-to-use Python API to define Large Language Models (LLMs) and support state-of-the-art optimizations to perform inference efficiently on NVIDIA GPUs. TensorRT-LLM also contains components to create Python and C++ runtimes that orchestrate the inference execution in performant way.

A machine learning compiler for GPUs, CPUs, and ML accelerators

Composable transformations of Python+NumPy programs: differentiate, vectorize, JIT to GPU/TPU, and more

MLX: An array framework for Apple silicon

No description.

🚀 Efficient implementations of state-of-the-art linear attention models in Torch and Triton

DOOM's fire algo, in zig, for 256 color terminals w/no dependencies

Fast and memory-efficient exact attention

No description.

Reference implementations of MLPerf™ training benchmarks

Callable PyTrees and filtered JIT/grad transformations => neural networks in JAX.

Safe Rust wrapper for the NVIDIA Management Library

Flax is a neural network library for JAX that is designed for flexibility.

Tensors and Dynamic neural networks in Python with strong GPU acceleration

Advanced Python Mastery (course by @dabeaz)

Preconditioned Conjugate Gradient in Pytorch

Implementation to convert a PyTorch UNet to Timeloop/Accelergy

A Python framework for high performance GPU simulation and graphics

A differentiable PDE solving framework for machine learning

Julia Wrappers for the arpack-ng Fortran library

A compiler for quantum computers based on A* and numerical optimization.

Run-time type checker for Python

:crab: Small exercises to get you used to reading and writing Rust code!

Master's Thesis for the MSc in Advanced Computer Science at Oxford University

A Python-level JIT compiler designed to make unmodified PyTorch programs faster.

functorch is JAX-like composable function transforms for PyTorch.

A youtube-dl fork with additional features and fixes

No description.

Implementation of R-GCNs for Relational Link Prediction

Rust native code for BQSKit

Small Python library to automatically set CUDA_VISIBLE_DEVICES to the least loaded device on multi-GPU systems.

A place to submit conda recipes before they become fully fledged conda-forge feedstocks

Python 3.6+ toolbox for submitting jobs to Slurm

Hydra is a framework for elegantly configuring complex applications

No description.

A conda-smithy repository for conda-forge-pinning.

A conda-smithy repository for dbgpy.

The lightweight PyTorch wrapper for high-performance AI research. Scale your models, not the boilerplate.

A library for programmatically generating equivariant layers through constraint solving

PointFlow : 3D Point Cloud Generation with Continuous Normalizing Flows

A Python3 NES emulator and OpenAI Gym interface

A simple tool for fluid simulation.

Pruning and other network surgery for trained Keras models.

A modular framework for neural networks with Euclidean symmetry

Einsum optimization using opt_einsum and PyTorch FX graph rewriting

Scripts and tidbits from work at CERN

The interface between ROOT and NumPy

Fast inference of Boosted Decision Trees in FPGAs

E(2)-Equivariant CNNs Library for Pytorch

Learn to write Rust procedural macros  [Rust Latam conference, Montevideo Uruguay, March 2019]

QKeras: a quantization deep learning library for Tensorflow Keras

No description.

Viewer for 4D shapes

No description.

Repository for bachelor project in mathematics

AI Hackathon / Equinor Software Innovation Trondheim / NTNU Brain

Implementations of ideas discussed in "Deep Learning as Optimal Control Problems: Models and Numerical Methods"

Deep Translation and Rotation Equivariance

Q-learning applied to a simple homemade racecar game.

A tool for speeding up reading.