Accelerate, Optimize performance with streamlined training and serving options with JAX.

Monocular whole-body 3D human pose estimation using the SOMA body model

easydel jax kernels writen in triton for gpus and pallas for tpus

🤗 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.

(EasyDel Former) is a utility library designed to simplify and enhance the development in JAX

EOpod is a streamlined command execution tool designed to run and manage operations on Google Cloud Pods efficiently

Fine-tuning & Reinforcement Learning for LLMs. 🦥 Train OpenAI gpt-oss, DeepSeek-R1, Qwen3, Gemma 3, TTS 2x faster with 70% less VRAM.

小说下载工具, 支持 TXT/EPUB/HTML 导出

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

Create Custom LLMs

Distributed Peer-to-Peer Web Search Engine and Intranet Search Appliance

Official inference code and LongText-Bench benchmark for our paper X-Omni (https://arxiv.org/pdf/2507.22058).

Official PyTorch implementation for Hogwild! Inference: Parallel LLM Generation with a Concurrent Attention Cache

Nano vLLM

封装OpenVPN，暴露 socks5 端口，支持 UDP 转发

NAS媒体库管理工具

一个为Emby、Jellyfin服务器提供Strm直链播放的小项目，推荐配合MediaWarp使用

A selfhosted Stremio addon

Converts OpenVPN connection to SOCKS5 proxy server in Docker

Wraps your program with OpenVPN network tunnel.

A self-hosted BitTorrent indexer, DHT crawler, content classifier and torrent search engine with web UI, GraphQL API and Servarr stack integration.

Replace OpenAI GPT with another LLM in your app by changing a single line of code. Xinference gives you the freedom to use any LLM you need. With Xinference, you're empowered to run inference with …