📚 从零开始的大语言模型原理与实践教程

方便好用的纯纯 C++ tokenizer 封装

Voice Activity Detector (VAD) : low-latency, high-performance and lightweight

Xiaomi Miloco

Depth Anything 3

ncnn version of CodeFormer

support yolov5/v8/11

[CVPR 2025] Official PyTorch implementation of "EdgeTAM: On-Device Track Anything Model"

RTMO demo on Axera NPU

New generation of CLIP with fine grained discrimination capability, ICML2025

Demo for PPOCRv5 on Axera device.

The Pipeline example based on AX650N/AX8850 AI Card, depend on AXCL API, shows the software development skills of Image Processing, NPU, Codec, and Display modules, which is helpful for users to de…

Demo for Qwen3 VL Series Models

ppocrv5(det, cls, rec) onnx/axmodel inference pipeline

molly, an LLM designed to understand multi-omics data.

Cosyvoice2 语音生成 DEMO

NVR with realtime local object detection for IP cameras

AXERA text embeding sdk base Qwen/Qwen3-Embedding-0.6B

As much as possible, facilitate users' assessment of the "One-Finger Zen" feature of the AX NPU, reduce the sunk cost, and encourage everyone to do something more valuable instead of getting stuck …

From Images to High-Fidelity 3D Assets with Production-Ready PBR Material

Text-to-Audio/Music Generation

Versatile audio super resolution (any -> 48kHz) with AudioSR.

LLM-powered framework for deep document understanding, semantic retrieval, and context-aware answers using RAG paradigm.