Radim Řehůřek

Biscuit. Biscuit – a high-performance, bitmap-based deterministic index for PostgreSQL

embedded-google-fonts. Embed (base64 encoded) google fonts into css.

tesseral. Open source auth infrastructure for B2B SaaS

llamafile. Distribute and run LLMs with a single file.

ramalama. RamaLama is an open-source developer tool that simplifies the local serving of AI models from any source and facilitates their use for inference in production, all through the familiar language of containers.

konfig. SDK & API Docs Generator. Sunset as of December 2024

lite-youtube-embed. A faster youtube embed.

YOLOX. YOLOX is a high-performance anchor-free YOLO, exceeding yolov3~v5 with MegEngine, ONNX, TensorRT, ncnn, and OpenVINO supported. Documentation: https://yolox.readthedocs.io/

open-parse. Improved file parsing for LLM’s

ocrs. Rust library and CLI tool for OCR (extracting text from images)

py-spy. Sampling profiler for Python programs

pybloomfiltermmap3. Fast Python Bloom Filter using Mmap

tensorli. Absolute minimalistic implementation of a GPT-like transformer using only numpy (<650 lines).

rehash. Resumable hashlib: a picklable interface to CPython's OpenSSL-based hashlib standard library

madbg. A fully-featured remote and preemptive debugger for python

USearch. Fast Open-Source Search & Clustering engine × for Vectors & Arbitrary Objects × in C++, C, Python, JavaScript, Rust, Java, Objective-C, Swift, C#, GoLang, and Wolfram 🔍

msoffcrypto-tool. Python tool and library for decrypting and encrypting MS Office files using passwords or other keys

python-docx. Create and modify Word documents with Python

autolabel. Label, clean and enrich text datasets with LLMs.

brainworkshop. Continued development of the popular brainworkshop game

ai-pdf-chatbot-langchain. AI PDF chatbot agent built with LangChain & LangGraph

Open-Assistant. OpenAssistant is a chat-based assistant that understands tasks, can interact with third-party systems, and retrieve information dynamically to do so.

indexed_gzip. Fast random access of gzip files in Python

langchain-visualizer. Visualization and debugging tool for LangChain workflows

ParallelZipFile. Quick hacky parallel ZipFile implementation

olefile. olefile is a Python package to parse, read and write Microsoft OLE2 files (also called Structured Storage, Compound File Binary Format or Compound Document File Format), such as Microsoft Office 97-2003 documents, vbaProject.bin in MS Office 2007+ files, Image Composer and FlashPix files, Outlook messages, StickyNotes, several Microscopy file formats, McAfee antivirus quarantine files, etc.

picoGPT. An unnecessarily tiny implementation of GPT-2 in NumPy.

keycloak. Open Source Identity and Access Management For Modern Applications and Services

hetzner-k3s. The easiest and fastest way to create production-ready Kubernetes clusters on Hetzner Cloud

unison. Unison file synchronizer

min-dalle. min(DALL·E) is a fast, minimal port of DALL·E Mini to PyTorch

bolt. 10x faster matrix and vector operations

pgvector. Open-source vector similarity search for Postgres

pgcompacttable. Perl

memray. Memray is a memory profiler for Python

personal-security-checklist. 🔒 A compiled checklist of 300+ tips for protecting digital security and privacy in 2026

python-xz. Pure Python implementation of the XZ file format with random access support

across. Action SuperCross Source Code

radix-sorting. Radix sorting from the ground up

mailcow-dockerized. mailcow: dockerized - 🐮 + 🐋 = 💕

textual. The lean application framework for Python. Build sophisticated user interfaces with a simple Python API. Run your apps in the terminal and a web browser.

static-ranges. WSGI middleware for handling byte-ranges

dive. A tool for exploring each layer in a docker image

matthewruffell.github.io. Personal website and blog

filprofiler. A Python memory profiler for data processing and scientific computing applications

polyleven. A Fast Levenshtein Distance Library for Python

TinyFastSS. An index data structure for approximate string search.

pg-costop. Vector Arithmetic and Weighted, Variably Randomized Cosine Similarity Search in Postgres

DALLE-pytorch. Implementation / replication of DALL-E, OpenAI's Text to Image Transformer, in Pytorch

Laplacians.jl. Algorithms inspired by graph Laplacians: linear equation solvers, sparsification, clustering, optimization, etc.

balbes. Teaching my kid to read in Russian

OCRmyPDF. OCRmyPDF adds an OCR text layer to scanned PDF files, allowing them to be searched

supertokens-core. Open source alternative to Auth0 / Firebase Auth / AWS Cognito

faster-cpython. How to make CPython faster.

kutuzov. Derives Python 3 type annotations from Sphinx docstrings

dangerzone. Take potentially dangerous PDFs, office documents, or images and convert them to safe PDFs

setuptools-cythonize. Distribute python modules/packages as binary files (compilation based on Cython)

Marketing-for-Engineers. A curated collection of marketing articles & tools to grow your product.

exchangelib. Python client for Microsoft Exchange Web Services (EWS)

FBGEMM. FB (Facebook) + GEMM (General Matrix-Matrix Multiplication) - https://code.fb.com/ml-applications/fbgemm/

pg_squeeze. A PostgreSQL extension for automatic bloat cleanup

pyphi. A toolbox for integrated information theory.

pdftagger. A tool to automatically tag existing pdfs

n2. TOROS N2 - lightweight approximate Nearest Neighbor library which runs fast even with large datasets

cChardet. universal character encoding detector

jsquery. JsQuery – json query language with GIN indexing support

interpret. Fit interpretable models. Explain blackbox machine learning.

smart_open. Utils for streaming large files (S3, HDFS, gzip, bz2...)

ai-reference-models. Intel® AI Reference Models: contains Intel optimizations for running deep learning workloads on Intel® Xeon® Scalable processors and Intel® Data Center GPUs

mars. Mars is a tensor-based unified framework for large-scale data computation which scales numpy, pandas, scikit-learn and Python functions.

pyahocorasick. Python module (C extension and plain python) implementing Aho-Corasick algorithm

TCN. Sequence modeling benchmarks and temporal convolutional networks

fairseq. Facebook AI Research Sequence-to-Sequence Toolkit written in Python.

pytext. A natural language modeling framework based on PyTorch

vaex. Out-of-Core hybrid Apache Arrow/NumPy DataFrame for Python, ML, visualization and exploration of big tabular data at a billion rows per second 🚀

NLP-progress. Repository to track the progress in Natural Language Processing (NLP), including the datasets and the current state-of-the-art for the most common NLP tasks.

bert. TensorFlow code and pre-trained models for BERT

py-aho-corasick. A pure Python implementation of Aho-Corasick algorithm.

NETL-Automatic-Topic-Labelling-. Generating labels for topics automatically using neural embeddings

flair. A very simple framework for state-of-the-art Natural Language Processing (NLP)

slackin. Public Slack organizations made easy

SmoothLife. Continuous Domain Game of Life in Python with Numpy

web2text. Source code for the paper "Web2Text: Deep Structured Boilerplate Removal", full paper @ ECIR'18

nlp-datasets. Alphabetical list of free/public domain datasets with text data for use in Natural Language Processing (NLP)

PyCEbox. ⬛ Python Individual Conditional Expectation Plot Toolbox

dlib. A toolkit for making real world machine learning and data analysis applications in C++

word2vec-explorer. Tool for exploring Word Vector models

umap. Uniform Manifold Approximation and Projection

gun. An open source cybersecurity protocol for syncing decentralized graph data.

VoxelSpace. Terrain rendering algorithm in less than 20 lines of code

gitalk. Gitalk is a modern comment component based on Github Issue and Preact.

elasticsearch-cloud-deploy. Deploy Elasticsearch on the cloud easily

gensim-data. Data repository for pretrained NLP models and NLP corpora.

Pyro4. Pyro 4.x - Python remote objects

streamz. Real-time stream processing for python

flashtext. Extract Keywords from sentence or Replace keywords in sentences.

deep-learning-with-python-notebooks. Jupyter notebooks for the code samples of the book "Deep Learning with Python"

textract. extract text from any document. no muss. no fuss.

spotlight. Deep recommender models using PyTorch.

corex_topic. Hierarchical unsupervised and semi-supervised topic models for sparse count data with CorEx

tagger. Named Entity Recognition Tool

simple-salesforce. A very simple Salesforce.com REST API client for Python

SentEval. A python tool for evaluating the quality of sentence embeddings.

NeuroNER. Named-entity recognition using neural networks. Easy-to-use and state-of-the-art results.

StackNet. StackNet is a computational, scalable and analytical Meta modelling framework

multiNLI. Python

nel. Entity linking framework

boltons. 🔩 Like builtins, but boltons. 250+ constructs, recipes, and snippets which extend (and rely on nothing but) the Python standard library. Nothing like Michael Bolton.

langid.py. Stand-alone language identification system

Zappa. Serverless Python

LargeVis. C++

word2gm. Word to Gaussian Mixture Model

cld3. C++

rasa. 💬 Open source machine learning framework to automate text- and voice-based conversations: NLU, dialogue management, connect to Slack, Facebook, and more - Create chatbots and voice assistants

tweet2vec_clustering. Python

yodaqa. A Question Answering system built on top of the Apache UIMA framework.

pysparnn. Approximate Nearest Neighbor Search for Sparse Data in Python!

LightLDA. Scalable, fast, and lightweight system for large-scale topic modeling

fairml. Python

pyrasite. Inject code into running Python processes

lopq. Training of Locally Optimized Product Quantization (LOPQ) models for approximate nearest neighbor search of high dimensional data in Python and Spark.

online-go.com. Source code for the Online-Go.com web interface

OgmaNeo. Ogma - OgmaNeo

pyminifier. Pyminifier is a Python code minifier, obfuscator, and compressor.

Neural-Dialogue-Generation. Lua

PyShortTextCategorization. Various Algorithms for Short Text Mining

DataflowJavaSDK. Google Cloud Dataflow provides a simple, powerful model for building both batch and streaming parallel data processing pipelines.

minpy. NumPy interface with mixed backend execution

dynet. DyNet: The Dynamic Neural Network Toolkit

big-list-of-naughty-strings. The Big List of Naughty Strings is a list of strings which have a high probability of causing issues when used as user-input data.

archived-desktop-app. The old electron based nteract notebook

stanford_dl_ex. Programming exercises for the Stanford Unsupervised Feature Learning and Deep Learning Tutorial

bazel. a fast, scalable, multi-language and extensible build system

ipex-llm. Accelerate local LLM inference and finetuning (LLaMA, Mistral, ChatGLM, Qwen, DeepSeek, Mixtral, Gemma, Phi, MiniCPM, Qwen-VL, MiniCPM-V, etc.) on Intel XPU (e.g., local PC with iGPU and NPU, discrete GPU such as Arc, Flex and Max); seamlessly integrate with llama.cpp, Ollama, HuggingFace, LangChain, LlamaIndex, vLLM, DeepSpeed, Axolotl, etc.

deep-learning-book. Repository for "Introduction to Artificial Neural Networks and Deep Learning: A Practical Guide with Applications in Python"

RankSys. Java 8 Recommender Systems framework for novelty, diversity and much more

git-of-theseus. Analyze how a Git repo grows over time

pyksvd. A highly optimized, parallel implementation of the Batch-OMP version of the KSVD learning algorithm.

LightGBM. A fast, distributed, high performance gradient boosting (GBT, GBDT, GBRT, GBM or MART) framework based on decision tree algorithms, used for ranking, classification and many other machine learning tasks.

FlexNeuART. Flexible classic and NeurAl Retrieval Toolkit

falcon. Chrome extension for full text history search!

fast-style-transfer. TensorFlow CNN for fast style transfer ⚡🖥🎨🖼

affinegap. :triangular_ruler: A Cython implementation of the affine gap string distance

pytextrank. Python implementation of TextRank algorithms ("textgraphs") for phrase extraction

berkeley-doc-summarizer. The Berkeley Document Summarizer is a learning-based, single-document summarization system that extracts source document content, exploits syntactic information to compress it, and uses coreference constraints to ensure clarity.

gifify. 😻 Convert any video file to an optimized animated GIF.

doc2text. Detect text blocks and OCR poorly scanned PDFs in bulk. Python module available via pip.

deep-learning-keras-tensorflow. Introduction to Deep Neural Networks with Keras and Tensorflow

deepmask. Torch implementation of DeepMask and SharpMask

ubuntu-ranking-dataset-creator. A script that creates train, valid and test datasets for the ranking task from Ubuntu corpus dialogs.

match. :crystal_ball: Scalable reverse image search built on Kubernetes and Elasticsearch

paragram-word. Python code for training Paragram word embeddings. These achieve human-level performance on some word similiarty tasks including SimLex-999.This code was used to obtain results in the appendix of our 2015 TACL paper "From Paraphrase Database to Compositional Paraphrase Model and Back".

Deep-Semantic-Similarity-Model. My Keras implementation of the Deep Semantic Similarity Model (DSSM)/Convolutional Latent Semantic Model (CLSM) described here: http://research.microsoft.com/pubs/226585/cikm2014_cdssm_final.pdf.

dataset-sts. Semantic Text Similarity Dataset Hub

charagram. Code to train and use models from "Charagram: Embedding Words and Sentences via Character n-grams".

keras-cam. Keras implementation of class activation mapping

ann-benchmarks. Benchmarks of approximate nearest neighbor libraries in Python

deep-simplification. Text simplification using RNNs

6.882-LDA. Implementation of Latent Dirichlet Allocation and application to DSPACE@MIT

curator. Curator: Tending your Elasticsearch indices

seq2seq. Sequence to Sequence Learning with Keras

qandabot. A conversational bot for assisting in customer support. Hackathon bot.

tpot. A Python Automated Machine Learning tool that optimizes machine learning pipelines using genetic programming.

rmn. relationship modeling networks (NAACL 2016)

darkforestGo. DarkForest, the Facebook Go engine.

gpq. A collection of tools for mining government data

tabula. Tabula is a tool for liberating data tables trapped inside PDF files

msg-extractor. Extracts emails and attachments saved in Microsoft Outlook's .msg files

vprof. Visual profiler for Python

tensorflow-glove. An implementation of GloVe in TensorFlow

DIY-Data-Science. Machine Learning Tool Guides and Theory Notes

deepdetect. Deep Learning Server and CLI for Torch and TensorRT

dynamic-means. A clustering library for large/temporally evolving datasets

dropout-plugin. Simple recommendation plugin for WordPress

spark-elastic. This project combines Apache Spark and Elasticsearch to enable mining & prediction for Elasticsearch.

histogram. Streaming Histograms for Clojure/Java

experimental-mf. cache-friendly multithread matrix factorization

nupic.subutai. Subutai's personal nupic code and experiments

DataflowPythonSDK. Google Cloud Dataflow provides a simple, powerful model for building both batch and streaming parallel data processing pipelines.

FALCONN. FAst Lookups of Cosine and Other Nearest Neighbors (based on fast locality-sensitive hashing)

sklearn-expertsys. Highly interpretable classifiers for scikit learn, producing easily understood decision rules instead of black box models

hug. Embrace the APIs of the future. Hug aims to make developing APIs as simple as possible, but no simpler.

osstracker. Github organization OSS metrics collector and metrics dashboard

airflow. Apache Airflow - A platform to programmatically author, schedule, and monitor workflows

weblas. GPU Powered BLAS for Browsers :gem:

char-rnn. Multi-layer Recurrent Neural Networks (LSTM, GRU, RNN) for character-level language models in Torch

lda2vec. Python

mlpack. mlpack: a fast, header-only C++ machine learning library

tapir. TAPIR distributed transactional storage system

atomiclong. A CFFI using AtomicLong type for CPython and PyPy.

visualize-text-reuse. Visualize text reuse with D3.js

marshmallow. A lightweight library for converting complex objects to and from simple Python datatypes.

fuzzysearch-demo. Demo of fuzzy (nearest neighbor) search with dense data

random-projections-at-berlinbuzzwords. Demo of random projections at BerlinBuzzwords 2015

SemanticSearchInNumpy. XSLT

data_science_delivered. Observations from Ian on successfully delivering data science products

d3-jupyter-tutorial. JavaScript

cloudpickle. Extended pickling support for Python objects

neural-vqa. :grey_question: Visual Question Answering in Torch

skflow. Simplified interface for TensorFlow (mimicking Scikit Learn) for Deep Learning