🤗 Transformers: the model-definition framework for state-of-the-art machine learning models in text, vision, audio, and multimodal models, for both inference and training.

FastAPI framework, high performance, easy to learn, fast to code, ready for production

🌐 Make websites accessible for AI agents. Automate tasks online with ease.

Web UI for training and running open models like Gemma 4, Qwen3.6, DeepSeek, gpt-oss locally.

DeepSpeed is a deep learning optimization library that makes distributed training and inference easy, efficient, and effective.

You like pytorch? You like micrograd? You love tinygrad! ❤️

Qwen3 is the large language model series developed by Qwen team, Alibaba Cloud.

Google Chromium, sans integration with Google

Build Real-Time Knowledge Graphs for AI Agents

[NeurIPS'23 Oral] Visual Instruction Tuning (LLaVA) built towards GPT-4V level capabilities and beyond.

Fast and memory-efficient exact attention

Fast and accurate AI powered file content types detection

Replace 'hub' with 'ingest' in any GitHub URL to get a prompt-friendly extract of a codebase

Supercharge Your LLM with the Fastest KV Cache Layer

On-device AI across mobile, embedded and edge for PyTorch

A compact implementation of SGLang, designed to demystify the complexities of modern LLM serving systems.

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

Minimalistic large language model 3D-parallelism training

Lighteval is your all-in-one toolkit for evaluating LLMs across multiple backends

cuTile is a programming model for writing parallel kernels for NVIDIA GPUs

A faithful reproduction of the "Attention is All You Need" paper's transformer architecture.

Parsit is a specialized vision-language model designed for document analysis tasks. Parsit combines Qwen3 language models with SigLIP-2 vision encoders to provide state-of-the-art performance on do…

A modular, block-by-block LLM building library

On-device AI across mobile, embedded and edge for PyTorch

A tool to convert audio (or video) files into spectrogram images using the Short Fast Fourier Transform (SFFT)

A Python library for extracting and classifying text elements from PDF documents. The processor uses advanced heuristics and machine learning techniques to identify different types of text elements…