Fast and memory-efficient exact attention

Qwen2.5-Omni is an end-to-end multimodal model by Qwen team at Alibaba Cloud, capable of understanding text, audio, vision, video, and performing real-time speech generation.

Automated nginx proxy for Docker containers using docker-gen

An open source implementation of CLIP.

This repo contains the code for "VLM2Vec: Training Vision-Language Models for Massive Multimodal Embedding Tasks" [ICLR 2025]

A file utility for accessing both local and remote files through a unified interface.

utilities for decoding deep representations (like sentence embeddings) back to text

CLIP-like model evaluation

PyTorch native post-training library

Ray is a unified framework for scaling AI and Python applications. Ray consists of a core distributed runtime and a toolkit of libraries (Ray AIR) for accelerating ML workloads.

Computing with Python functions.

Python Packaging User Guide

Code for BLT research paper

Command line driven CI frontend and development task automation tool.

Multiple dispatch

scikit-learn: machine learning in Python

Re-Connectable secure remote shell

A conda-smithy repository for lpips.

MTEB: Massive Text Embedding Benchmark

Fork of webdataset

Large-scale text-video dataset.

Python disk-backed cache (Django-compatible). Faster than Redis and Memcached. Pure-Python.

PyTorch code accompanies the TVQA dataset paper, in EMNLP 2018

⏰ AI conference deadline countdowns