build-once run-anywhere c library

a pptx to markdown converter

4M: Massively Multimodal Masked Modeling

PygmalionAI's large-scale inference engine

Convert Compute And Books Into Instruct-Tuning Datasets

NeMo Guardrails is an open-source toolkit for easily adding programmable guardrails to LLM-based conversational systems.

StreamSpeech is an “All in One” seamless model for offline and simultaneous speech recognition, speech translation and speech synthesis.

Explorations into the proposal from the paper "Grokfast, Accelerated Grokking by Amplifying Slow Gradients"

The NVIDIA RTX™ AI Toolkit is a suite of tools and SDKs for Windows developers to customize, optimize, and deploy AI models across RTX PCs and cloud.

x86-64 code/pe virtualizer

Technically-oriented PDF Collection (Papers, Specs, Decks, Manuals, etc)

Tiny ASIC implementation for "The Era of 1-bit LLMs All Large Language Models are in 1.58 Bits" matrix multiplication unit

Code for Zero-Shot Tokenizer Transfer

Reproduce CKA: Similarity of Neural Network Representations Revisited

An Extensible Deep Learning Library

Find the Best LLM for Your Needs through E2E Testing

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

TunnlTo is a Windows WireGuard VPN client built for split tunnelling.

MARS5 speech model (TTS) from CAMB.AI

Implementation for MatMul-free LM.

Doing simple retrieval from LLM models at various context lengths to measure accuracy

Like "ls", but for images. Shows thumbnails in terminal using sixel graphics.

GPUOcelot: A dynamic compilation framework for PTX

⚡FlashRAG: A Python Toolkit for Efficient RAG Research

Pytorch implementation of a simple way to enable (Stochastic) Frame Averaging for any network

a research paper for generative cartoon interpolation

Generative neural networks for 3D terrain.