TinyMaix is a tiny inference library for microcontrollers (TinyML).

Mobile-Agent: The Powerful GUI Agent Family

Python tool for converting files and office documents to Markdown.

The all-in-one AI productivity accelerator. On device and privacy first with no annoying setup or configuration.

✨✨[NeurIPS 2025] VITA-1.5: Towards GPT-4o Level Real-Time Vision and Speech Interaction

YOLOv5 🚀 in PyTorch > ONNX > CoreML > TFLite

C++ based gRPC (C++, Python, Ruby, Objective-C, PHP, C#)

Privacy-Preserving Computing Platform 由密码学专家团队打造的开源隐私计算平台，支持多方安全计算、联邦学习、隐私求交、匿踪查询等。

S-LoRA: Serving Thousands of Concurrent LoRA Adapters

Use PEFT or Full-parameter to CPT/SFT/DPO/GRPO 600+ LLMs (Qwen3.6, DeepSeek-R1, GLM-5.1, InternLM3, Llama4, ...) and 300+ MLLMs (Qwen3-VL, Qwen3-Omni, InternVL3.5, Ovis2.5, GLM4.5v, Gemma4, Llava, …

Open-source, accurate and easy-to-use video speech recognition & clipping tool, LLM based AI clipping intergrated.

A Repository for Single- and Multi-modal Speaker Verification, Speaker Recognition and Speaker Diarization

For releasing code related to compression methods for transformers, accompanying our publications

System for AI Education Resource.

pycorrector is a toolkit for text error correction. 文本纠错，实现了Kenlm，T5，MacBERT，ChatGLM3，Qwen2.5等模型应用在纠错场景，开箱即用。

A generative speech model for daily dialogue.

Easy to use open source fast database for search | Good alternative to Elasticsearch | Drop-in replacement for E in the ELK stack

TinyChatEngine: On-Device LLM Inference Library

Offsite-Tuning: Transfer Learning without Full Model

An open source implementation of Microsoft's VALL-E X zero-shot TTS model. Demo is available in https://plachtaa.github.io/vallex/

This repo is a pipeline of VITS finetuning for fast speaker adaptation TTS, and many-to-many voice conversion

"结巴"中文分词的C++版本

C++ websocket client/server library

💯2026年 系统架构设计师（软考高级）备考资源库+配套免费刷题软件。

Federated Learning for LLMs.

[ACL 2020] Generating Diverse and Consistent QA pairs from Contexts with Information-Maximizing Hierarchical Conditional VAEs

fay是一个帮助数字人（2.5d、3d、移动、pc、网页）或大语言模型（openai兼容、deepseek）连通业务系统的agent框架。

An easy-to-use federated learning platform

An Industrial Grade Federated Learning Framework