Ecosystem & Stack: paddlepaddle

A high-performance inference and deployment toolkit for Large Language Models (LLMs) and Vision-Language Models (VLMs) based on PaddlePaddle.

An easy-to-use open-source toolkit built on PaddlePaddle, offering state-of-the-art models for diverse speech and audio tasks like ASR, TTS, translation, and speaker verification.

A high-performance, comprehensive face recognition library built on PaddlePaddle and PyTorch.