Anchored Preference Optimization and Contrastive Revisions: Addressing Underspecification in Alignment

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

SearXNG is a free internet metasearch engine which aggregates results from various search services and databases. Users are neither tracked nor profiled.

⬛️ CLI tool for saving complete web pages as a single HTML file

We write your reusable computer vision tools. 💜

A distributed, fault-tolerant task queue

Easily embed, cluster and semantically label text datasets

Unified Structure Generation for Universal Information Extraction

A flexible package for multimodal-deep-learning to combine tabular data with text and images using Wide and Deep models in Pytorch

Generative Representational Instruction Tuning

Tantivy is a full-text search engine library inspired by Apache Lucene and written in Rust

Repair invalid JSON documents

the portable Python dataframe library

A configuration as code language with rich validation and tooling.

Inquisitive Parrots for Search

A simple toolkit to process TREC files in Python.

highlight.io: The open source, full-stack monitoring platform. Error monitoring, session replay, logging, distributed tracing, and more.

OCR, layout analysis, reading order, table recognition in 90+ languages

Spark library for generalized K-Means clustering. Supports general Bregman divergences. Suitable for clustering probabilistic data, time series data, high dimensional data, and very large data.

YuLan-IR: Information Retrieval Boosted LMs

Stop messing around with finicky sampling parameters and just use DRµGS!

An open source, non-profit web search engine

Geziyor, blazing fast web crawling & scraping framework for Go. Supports JS rendering.

Machine Learning Engineering Open Book

Given a scholarly PDF, extract figures, tables, captions, and section titles.

Configuration Management for Python ⚙

🔥 🔥 🔥 Open Source JIRA, Linear, Monday, and Asana Alternative. Plane helps you track your issues, epics, and product roadmaps in the simplest way possible.

🦦 Otter, a multi-modal model based on OpenFlamingo (open-sourced version of DeepMind's Flamingo), trained on MIMIC-IT and showcasing improved instruction-following and in-context learning ability.

Simple, efficient background processing for Ruby

Stealth browsers as a service. Connect your scraper or automation to a fleet of cloud-hosted browsers configured for reliability and stealth.