Public repository for Agent Skills

A high-throughput and memory-efficient inference and serving engine for LLMs

A family of state-of-the-art Transformer-based audio codecs for low-bitrate high-quality audio coding.

The AI Scientist: Towards Fully Automated Open-Ended Scientific Discovery 🧑‍🔬

[NeurIPS 2024] Depth Anything V2. A More Capable Foundation Model for Monocular Depth Estimation

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

An extremely fast Python linter and code formatter, written in Rust.

Development repository for the Triton language and compiler

Fast, flexible LLM inference

[CVPR 2024] Depth Anything: Unleashing the Power of Large-Scale Unlabeled Data. Foundation Model for Monocular Depth Estimation

ONNX-compatible Depth Anything: Unleashing the Power of Large-Scale Unlabeled Data

[ICLR 2025] HD-Painter: High-Resolution and Prompt-Faithful Text-Guided Image Inpainting with Diffusion Models

[AAAI 2024] VSFormer: Visual-Spatial Fusion Transformer for Correspondence Pruning

ONNX-compatible DocShadow: High-Resolution Document Shadow Removal. Supports TensorRT 🚀

[ICCV 2023] A large-scale high-resolution dataset satisfies all important data features about document shadow, covers a large number of document shadow images.

Focus on prompting and generating

ONNX-compatible Fast SeamlessM4T—Massively Multilingual & Multimodal Machine Translation

LightGlue: Local Feature Matching at Light Speed (ICCV 2023)

[3DV 2024 Oral] DeDoDe 🎶 Detect, Don't Describe --- Describe, Don't Detect, for Local Feature Matching

A repository for storing models that have been inter-converted between various frameworks. Supported frameworks are TensorFlow, PyTorch, ONNX, OpenVINO, TFJS, TFTRT, TensorFlowLite (Float32/16/INT8…

ONNX-compatible DeDoDe 🎶 Detect, Don't Describe - Describe, Don't Detect, for Local Feature Matching. Supports TensorRT 🚀

Official implementation of "Separate Anything You Describe"

Fast and memory-efficient exact attention

Tensors and Dynamic neural networks in Python with strong GPU acceleration

🤗 Diffusers: State-of-the-art diffusion models for image, video, and audio generation in PyTorch.

🤗 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.

ONNX-compatible LightGlue: Local Feature Matching at Light Speed. Supports TensorRT, OpenVINO