nwdxlgzs
Follow
Starred repositories
An accurate GUI element detection approach based on old-fashioned CV algorithms [Upgraded on 5/July/2021]
Ultralytics YOLO with Additional Knowledge Distillation Capability
Self-Created Tools to convert ONNX files (NCHW) to TensorFlow/TFLite/Keras format (NHWC). The purpose of this tool is to solve the massive Transpose extrapolation problem in onnx-tensorflow (onnx-t…
A modern GUI client based on Tauri, designed to run in Windows, macOS and Linux for tailored proxy experience
高性能OCR库，有PaddleOCR v4/v5模型。支持文本检测/识别、多语言（中文/英文/日文)。 提供Rust库 + CAPI动态库 + CLI工具，轻松集成 调用简单 开箱即用。 High-performance OCR library, supports multiple languages (Chinese/English/Japanese), provides Rust c…
opencv onnxruntime ocr android demo, jetpack compose + kotlin
The official repository for ERNIE 4.5 and ERNIEKit – its industrial-grade development toolkit based on PaddlePaddle.
chat log tool, easily use your own chat data. 聊天记录工具，轻松使用自己的聊天数据
Tool for onnx->keras or onnx->tflite. Hope this tool can help you.
ReLE中文大模型能力评测（持续更新）：目前已囊括257个大模型，覆盖chatgpt、gpt-4.1、o4-mini、谷歌gemini-2.5、Claude、智谱GLM-Z1、文心一言、qwen-max、百川、讯飞星火、商汤senseChat、minimax等商用模型， 以及DeepSeek-R1-0528、qwq-32b、deepseek-v3、qwen3、llama4、phi-4、glm…
A collection of 🤗 Transformers.js demos and example applications
High-performance In-browser LLM Inference Engine
Sample code for how to use Android binders from native (C++) space.
System V shared memory emulation on Android, using ashmem. That includes shmget, shmat, shmdt and shmctl functions.
ELFIO is a header-only C++ library for reading and generating ELF (Executable and Linkable Format) files.
Load elf file on HarmonyOS. Using https://github.com/MikhailProg/elf as elf loader.
LiteRT continues the legacy of TensorFlow Lite as the trusted, high-performance runtime for on-device AI. Now with LiteRT Next, we're expanding our vision with a new generation of APIs designed for…
A gallery that showcases on-device ML/GenAI use cases and allows people to try and use models locally.
Use PEFT or Full-parameter to CPT/SFT/DPO/GRPO 500+ LLMs (Qwen3, Qwen3-MoE, Llama4, GLM4.5, InternLM3, DeepSeek-R1, ...) and 200+ MLLMs (Qwen2.5-VL, Qwen2.5-Omni, Qwen2-Audio, Ovis2, InternVL3, Lla…
Unified Efficient Fine-Tuning of 100+ LLMs & VLMs (ACL 2024)
High-performance Inference and Deployment Toolkit for LLMs and VLMs based on PaddlePaddle