Python tool for converting files and office documents to Markdown.

A CLI tool for managing OpenAI batch processing jobs with ease.

Retrieval and Retrieval-augmented LLMs

Binary Passage Retriever (BPR) - an efficient passage retriever for open-domain question answering

EMNLP 2021 - Pre-training architectures for dense retrieval

Codebase for RetroMAE and beyond.

⚡ boost inference speed of T5 models by 5x & reduce the model size by 3x.

Late Interaction Models Training & Retrieval

TextGrad: Automatic ''Differentiation'' via Text -- using large language models to backpropagate textual gradients.

Tiny client for LLMs with vision and tool calling. As simple as it gets.

An LLM-powered knowledge curation system that researches a topic and generates a full-length report with citations.

Full text search in your Pandas dataframe

A simple, performant and scalable Jax LLM!

Library for reading and processing ML training data.

A lightweight, low-dependency, unified API to use all common reranking and cross-encoder models.

The prime repository for state-of-the-art Multilingual Question Answering research and development.

Run PyTorch models in the browser using ONNX.js

A pair of tiny foundational models trained in Brazilian Portuguese.🦙🦙

Code for experiments with transfer learning (on deep neural language models) between languages.

Rax is a Learning-to-Rank library written in JAX.

🌠 Manage your shell commands.

Run Effective Large Batch Contrastive Learning Beyond GPU/TPU Memory Constraint

Minimalistic large language model 3D-parallelism training

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