ashishdatta

gradient_descent_viz

interactive visualization of 5 popular gradient descent methods with step-by-step illustration and hyperparameter tuning UI

hardtime.nvim

A Neovim plugin helping you establish good command workflow and quit bad habit

llm.c

LLM training in simple, raw C/CUDA

data-is-better-together

Let's build better datasets, together!

DiskANN

Graph-structured Indices for Scalable, Fast, Fresh and Filtered Approximate Nearest Neighbor Search

NeMo-Curator

Scalable toolkit for data curation

StringZilla

Up to 10x faster strings for C, C++, Python, Rust, and Swift, leveraging SWAR and SIMD on Arm Neon and x86 AVX2 & AVX-512-capable chips to accelerate search, sort, edit distances, alignment scores, etc 🦖

hyperfine

A command-line benchmarking tool

gpt-fast

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

submitit

Python 3.8+ toolbox for submitting jobs to Slurm

tracy

Frame profiler

datatrove

Freeing data processing from scripting madness by providing a set of platform-agnostic customizable pipeline processing blocks.

lm-evaluation-harness

A framework for few-shot evaluation of language models.

mlx-examples

Examples in the MLX framework

mlx

MLX: An array framework for Apple silicon

llama.cpp

LLM inference in C/C++

stb

stb single-file public domain libraries for C/C++

mamba-minimal

Simple, minimal implementation of the Mamba SSM in one file of PyTorch.

mamba

Mamba SSM architecture

miller

Miller is like awk, sed, cut, join, and sort for name-indexed data such as CSV, TSV, and tabular JSON

tokenize-anything

Tokenize Anything via Prompting

CogVLM

a state-of-the-art-level open visual language model | 多模态预训练模型

tree-sitter

An incremental parsing system for programming tools

dotfiles

YADR - The best vim,git,zsh plugins and the cleanest vimrc you've ever seen

llm-awq

[MLSys 2024 Best Paper Award] AWQ: Activation-aware Weight Quantization for LLM Compression and Acceleration

AutoGPTQ

An easy-to-use LLMs quantization package with user-friendly apis, based on GPTQ algorithm.

gpt4all

gpt4all: run open-source LLMs anywhere

ollama

Get up and running with Llama 3, Mistral, Gemma, and other large language models.

megablocks

ruff-lsp

A Language Server Protocol implementation for Ruff.