GitLab

Diffusers is the go-to library for state-of-the-art pretrained diffusion models for generating images, audio, and even 3D structures of molecules.

Qwen2-Audio能够接受各种音频信号输入，并根据语音指令执行音频分析或直接响应文本。

phi-4 是一个 140 亿参数的模型，基于 Transformer 架构，专注于提升推理和 STEM 领域的问答能力。

The most powerful and modular stable diffusion GUI, api and backend with a graph/nodes interface.

unsloth框架基于triton优化模型训练速度和显存占用，使用Unsloth微调Mistral、Gemma、Llama时,速度可提高2-5倍,内存使用可减少70%!

VibeVoice 是一个新颖的框架，旨在从文本生成富有表现力的长篇多说话人对话音频，例如播客。

🤗 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.

RepViT在iPhone 12上以1ms的延迟实现了超过80%的top-1准确率，本算法基于RepViT进一步优化后准确率超过82%。

An open-source computational fluid dynamics software accelerated by GPU called OpenCFD-SCU for direct numerical simulation of compressible turbulence. 一款使用了GPU加速主要用于可压缩湍流直接数值模拟的开源软件，OpenCFD-SCU。

Personal TTS，即个性化语音合成，以阿里的KAN-TTS框架实现