yaysummeriscoming

deepsparse. Sparsity-aware deep learning inference runtime for CPUs

flash-tokenizer. EFFICIENT AND OPTIMIZED TOKENIZER ENGINE FOR LLM INFERENCE SERVING

dynamo. A Datacenter Scale Distributed Inference Serving Framework

EarTrumpet. EarTrumpet - Volume Control for Windows

DeepSeek-R1.

RapidOCR. 📄 Awesome OCR multiple programing languages toolkits based on ONNX Runtime, OpenVINO, MNN, PaddlePaddle, TensorRT and PyTorch.

gemlite. Fast low-bit matmul kernels in Triton

yalm. Yet Another Language Model: LLM inference in C++/CUDA, no libraries except for I/O

AlDente-Battery_Care_and_Monitoring. Menubar Tool to set Charge Limits and Prolong Battery Lifespan

Evaluate-Liger-Unsloth. Jupyter Notebook

SimPO. [NeurIPS 2024] SimPO: Simple Preference Optimization with a Reference-Free Reward

Liger-Kernel. Efficient Triton Kernels for LLM Training

alignment-handbook. Robust recipes to align language models with human and AI preferences

OnnxTR. OnnxTR a docTR (Document Text Recognition) library Onnx pipeline wrapper - for seamless, high-performing & accessible OCR

PiSSA. PiSSA: Principal Singular Values and Singular Vectors Adaptation of Large Language Models(NeurIPS 2024 Spotlight)

MultimodalOCR. On the Hidden Mystery of OCR in Large Multimodal Models (OCRBench)

AirPodsSanity. Fixes the Sound Quality of AirPods when connected to a Mac.

EETQ. Easy and Efficient Quantization for Transformers

augmentoolkit. Create Custom LLMs

optuna. A hyperparameter optimization framework

mistral.rs. Fast, flexible LLM inference

remote-faster-whisper. A basic HTTP API for handling Faster Whisper audio transcriptions over the network

ovos-stt-plugin-fasterwhisper. FasterWhisper for OpenVoiceOS

HALOs. A library with extensible implementations of DPO, KTO, PPO, ORPO, and other human-aware loss functions (HALOs).

insanely-fast-whisper. Jupyter Notebook

SqueezeLLM. [ICML 2024] SqueezeLLM: Dense-and-Sparse Quantization

TensorRT-LLM. TensorRT LLM provides users with an easy-to-use Python API to define Large Language Models (LLMs) and supports state-of-the-art optimizations to perform inference efficiently on NVIDIA GPUs. TensorRT LLM also contains components to create Python and C++ runtimes that orchestrate the inference execution in a performant way.

InternLM. Official release of InternLM series (InternLM, InternLM2, InternLM2.5, InternLM3).

Platypus. Code for fine-tuning Platypus fam LLMs using LoRA

trl. Train transformer language models with reinforcement learning.

AutoGPTQ. An easy-to-use LLMs quantization package with user-friendly apis, based on GPTQ algorithm.

ggml. Tensor library for machine learning

x-transformers. A concise but complete full-attention transformer with a set of promising experimental features from various papers

ColossalAI. Making large AI models cheaper, faster and more accessible

OCRmyPDF. OCRmyPDF adds an OCR text layer to scanned PDF files, allowing them to be searched

Promptify. Prompt Engineering | Prompt Versioning | Use GPT or other prompt based models to get structured output. Join our discord for Prompt-Engineering, LLMs and other latest research

faster-whisper. Faster Whisper transcription with CTranslate2

pytriton. PyTriton is a Flask/FastAPI-like interface that simplifies Triton's deployment in Python environments.

onnx-modifier. A tool to modify ONNX models in a visualization fashion, based on Netron and Flask.

vllm. A high-throughput and memory-efficient inference and serving engine for LLMs

nlpaug. Data augmentation for NLP

tuning_playbook. A playbook for systematically maximizing the performance of deep learning models.

VIMER. 视觉预训练基础模型仓库

pai-thin-client. A python client used to interact with the Private AI's API

RWKV-LM. RWKV (pronounced RwaKuv) is an RNN with great LLM performance, which can also be directly trained like a GPT transformer (parallelizable). We are at RWKV-7 "Goose". So it's combining the best of RNN and transformer - great performance, linear time, constant space (no kv-cache), fast training, infinite ctx_len, and free sentence embedding.

whisper. Robust Speech Recognition via Large-Scale Weak Supervision

doctr. docTR (Document Text Recognition) - a seamless, high-performing & accessible library for OCR-related tasks powered by Deep Learning. Ongoing development and maintenance by t2k.

ERNIE-Layout-Pytorch. An unofficial Pytorch implementation of ERNIE-Layout which is originally released through PaddleNLP.

parseq. Scene Text Recognition with Permuted Autoregressive Sequence Models (ECCV 2022)

peft. 🤗 PEFT: State-of-the-art Parameter-Efficient Fine-Tuning.

whisperX. WhisperX: Automatic Speech Recognition with Word-level Timestamps (& Diarization)

lingua-py. The most accurate natural language detection library for Python, suitable for short text and mixed-language text

openai-python. The official Python library for the OpenAI API

CLIP. CLIP (Contrastive Language-Image Pretraining), Predict the most relevant text snippet given an image

container-diff. container-diff: Diff your Docker containers

argilla. Argilla is a collaboration tool for AI engineers and domain experts to build high-quality datasets

ppocr-onnx. 利用 onnxruntime 及 PaddleOCR 提供的模型, 对图片中的文字进行检测与识别.

fastertransformer_backend. Python

TrustAI. 飞桨可信AI

mmocr. OpenMMLab Text Detection, Recognition and Understanding Toolbox

whisper-asr-webservice. OpenAI Whisper ASR Webservice API

unilm. Large-scale Self-supervised Pre-training Across Tasks, Languages, and Modalities

detoxify. Trained models & code to predict toxic comments on all 3 Jigsaw Toxic Comment Challenges. Built using ⚡ Pytorch Lightning and 🤗 Transformers. For access to our API, please email us at contact@unitary.ai.

TextAttack. TextAttack 🐙 is a Python framework for adversarial attacks, data augmentation, and model training in NLP https://textattack.readthedocs.io/en/master/

BlackHole. BlackHole is a modern macOS audio loopback driver that allows applications to pass audio to other applications with zero additional latency.

fastapi. FastAPI framework, high performance, easy to learn, fast to code, ready for production

dolt. Dolt – Git for Data

FasterTransformer. Transformer related optimization, including BERT, GPT

BERN2. BERN2: an advanced neural biomedical namedentity recognition and normalization tool

mmdetection. OpenMMLab Detection Toolbox and Benchmark

DeBERTa. The implementation of DeBERTa

ecco. Explain, analyze, and visualize NLP language models. Ecco creates interactive visualizations directly in Jupyter notebooks explaining the behavior of Transformer-based language models (like GPT2, BERT, RoBERTA, T5, and T0).

deid-examples. Examples scripts that showcase how to use Private AI Text to de-identify, redact, hash, tokenize, mask and synthesize PII in text.

transformer-deploy. Efficient, scalable and enterprise-grade CPU/GPU inference server for 🤗 Hugging Face transformer models 🚀

python-docx2txt. A pure python based utility to extract text and images from docx files.

grype. A vulnerability scanner for container images and filesystems

pdfrw. pdfrw is a pure Python library that reads and writes PDFs

PyPDF4. A utility to read and write PDFs with Python

PyPDF3. A utility to read and write PDFs with Python

textract. extract text from any document. no muss. no fuss.

python-docx. Create and modify Word documents with Python

mistral. Mistral: A strong, northwesterly wind: Framework for transparent and accessible large-scale language model training, built with Hugging Face 🤗 Transformers.

client. Triton Python, C++ and Java client libraries, and GRPC-generated client examples for go, java and scala.

nni. An open source AutoML toolkit for automate machine learning lifecycle, including feature engineering, neural architecture search, model compression and hyper-parameter tuning.

deepdetect. Deep Learning Server and CLI for Torch and TensorRT

uvicorn. An ASGI web server, for Python. 🦄

pybind11. Seamless operability between C++11 and Python

DNN-bench. A library that lets you easily increase efficiency of your deep learning models with no loss of accuracy.

xtreme. XTREME is a benchmark for the evaluation of the cross-lingual generalization ability of pre-trained multilingual models that covers 40 typologically diverse languages and includes nine tasks.

onnxruntime. ONNX Runtime: cross-platform, high performance ML inferencing and training accelerator

onnx-tensorrt. ONNX-TensorRT: TensorRT backend for ONNX

datasets. 🤗 The largest hub of ready-to-use NLP datasets for ML models with fast, easy-to-use and efficient data manipulation tools

server. The Triton Inference Server provides an optimized cloud and edge inferencing solution.

TurboTransformers. a fast and user-friendly runtime for transformer inference (Bert, Albert, GPT2, Decoders, etc) on CPU and GPU.

flask-healthz. Liveness and readiness endpoints for Flask

flair. A very simple framework for state-of-the-art Natural Language Processing (NLP)

doccano. Open source annotation tool for machine learning practitioners.

tensorflow_macos. TensorFlow for macOS 11.0+ accelerated using Apple's ML Compute framework.

CV-Parsing-using-Spacy-3. Resume and CV Summarization and Paring with Spacy in Python

name-dataset. The Python library for names.

layout-parser. A Unified Toolkit for Deep Learning Based Document Image Analysis

the-pile. Python

PaddleOCR. Turn any PDF or image document into structured data for your AI. A powerful, lightweight OCR toolkit that bridges the gap between images/PDFs and LLMs. Supports 100+ languages.

EasyOCR. Ready-to-use OCR with 80+ supported languages and all popular writing scripts including Latin, Chinese, Arabic, Devanagari, Cyrillic and etc.

yagmail. Send email in Python conveniently for gmail using yagmail

norfair. Lightweight Python library for adding real-time multi-object tracking to any detector.

ML-with-Android-11. A repository demonstrating all that's new in Android 11 for ML and how you could try it out for your own use-cases

torch-blocksparse. Block-sparse primitives for PyTorch

datasets. 🤗 The largest hub of ready-to-use datasets for AI models with fast, easy-to-use and efficient data manipulation tools

huggingface_libtorch. Minimal example of using a traced huggingface transformers model with libtorch

openwebtext. Open clone of OpenAI's unreleased WebText dataset scraper. This version uses pushshift.io files instead of the API for speed.

cc_net. Tools to download and cleanup Common Crawl data

tokenizers. 💥 Fast State-of-the-Art Tokenizers optimized for Research and Production

onnxt5. Summarization, translation, sentiment-analysis, text-generation and more at blazing speed using a T5 version implemented in ONNX.

fast-reid. SOTA Re-identification Methods and Toolbox

bert_extension_tf. BERT Extension in TensorFlow

gan-compression. [CVPR 2020] GAN Compression: Efficient Architectures for Interactive Conditional GANs

torchprofile. Count the MACs / FLOPs of PyTorch models

tensorrtx. Implementation of popular deep learning networks with TensorRT network definition API

Pytorch_Retinaface. Retinaface get 80.99% in widerface hard val using mobilenet0.25.

selfieBenchmark. MATLAB

insightface. State-of-the-art 2D and 3D Face Analysis Project

google-research. Google Research

transformers. 🤗 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.

darknet. YOLOv4 / Scaled-YOLOv4 / YOLO - Neural Networks for Object Detection (Windows and Linux version of Darknet )

first-order-model. This repository contains the source code for the paper First Order Motion Model for Image Animation

Swift-Regex-Tester. Small macOS app to test regex in Swift.

python-phonenumbers. Python port of Google's libphonenumber

keras-ocr. A packaged and flexible version of the CRAFT text detector and Keras CRNN recognition model.

AdelaiDet. AdelaiDet is an open source toolbox for multiple instance-level detection and recognition tasks.

quickstart-android. Firebase Quickstart Samples for Android

Apple_VS_Google. Comparing Apple's and Google's on-device OCR technologies

keras-CenterNet. CenterNet (Objects as Points) implementation in Keras and Tensorflow

CenterNet-tensorflow. CenterNet: Objects as Points in Tensorflow

netron. Visualizer for neural network, deep learning and machine learning models

swiftshield. 🔒 Swift Obfuscator that protects iOS apps against reverse engineering attacks.

bert. TensorFlow code and pre-trained models for BERT

ShuffleNet-Series. Python

swift-coreml-transformers. Swift Core ML 3 implementations of GPT-2, DistilGPT-2, BERT, and DistilBERT for Question answering. Other Transformers coming soon!

tf-coreml. TensorFlow to CoreML Converter

radish. C++ model train&inference framework

tpu. Reference models and tools for Cloud TPUs.

gen-efficientnet-pytorch. Pretrained EfficientNet, EfficientNet-Lite, MixNet, MobileNetV3 / V2, MNASNet A1 and B1, FBNet, Single-Path NAS

DeepLearningExamples. State-of-the-Art Deep Learning scripts organized by models - easy to train and deploy with reproducible accuracy and performance on enterprise-grade infrastructure.

DataLoaders_DALI. PyTorch DataLoaders implemented with DALI for accelerating image preprocessing

tfjs. A WebGL accelerated JavaScript library for training and deploying ML models.

XLM. PyTorch original implementation of Cross-lingual Language Model Pretraining.

XNNPACK. High-efficiency floating-point neural network inference operators for mobile, server, and Web

Object-Detection-Metrics. Most popular metrics used to evaluate object detection algorithms.

webml-polyfill. Deprecated, the Web Neural Network Polyfill project has been moved to https://github.com/webmachinelearning/webnn-polyfill

tfrecord-view. Consume and view contents of TFRecord files (for Tensorflow's object detection api)

tensorflow-onnx. Convert TensorFlow, Keras, Tensorflow.js and Tflite models to ONNX

pytorch-ssd. MobileNetV1, MobileNetV2, VGG based SSD/SSD-lite implementation in Pytorch 1.0 / Pytorch 0.4. Out-of-box support for retraining on Open Images dataset. ONNX and Caffe2 support. Experiment Ideas like CoordConv.

pytorch-image-models. The largest collection of PyTorch image encoders / backbones. Including train, eval, inference, export scripts, and pretrained weights -- ResNet, ResNeXT, EfficientNet, NFNet, Vision Transformer (ViT), MobileNetV4, MobileNet-V3 & V2, RegNet, DPN, CSPNet, Swin Transformer, MaxViT, CoAtNet, ConvNeXt, and more

pytorch-mobilenet-v3. MobileNetV3 in pytorch and ImageNet pretrained models

hiddenlayer. Neural network graphs and training metrics for PyTorch, Tensorflow, and Keras.

kann. A lightweight C library for artificial neural networks

CondenseNet. CondenseNet: Light weighted CNN for mobile devices

models. Models and examples built with TensorFlow

pytorch-mobilenet-v2. A PyTorch implementation of MobileNet V2 architecture and pretrained model.

arxiv-network-graph. Arxiv's ML papers network graph and browser

convnet-burden. Memory consumption and FLOP count estimates for convnets

Shufflenet-v2-Pytorch. Perfect implement. Model shared. x0.5 (Top1:60.646) and 1.0x (Top1:69.402).

xla. Enabling PyTorch on XLA Devices (e.g. Google TPU)

fastai_wd. Jupyter Notebook

pytorch. Tensors and Dynamic neural networks in Python with strong GPU acceleration

kindling. A Swiss Army knife for building dope stuff with pytorch.

opencv_transforms. OpenCV implementation of Torchvision's image augmentations

knowledge-distillation-pytorch. A PyTorch implementation for exploring deep and shallow knowledge distillation (KD) experiments with flexibility

NCMN. A PyTorch implementation of non-correlating multiplicative noise (NCMN).

backdrop. Implementation and demonstration of backdrop in pytorch. Code and demonstration of GP dataset generator.

DALI. A GPU-accelerated library containing highly optimized building blocks and an execution engine for data processing to accelerate deep learning training and inference applications.

AdamW-pytorch. Implementation and experiments for AdamW on Pytorch

cifar2png. Convert CIFAR-10 and CIFAR-100 datasets into PNG images

apex. A PyTorch Extension: Tools for easy mixed precision and distributed training in Pytorch

distillation. Keras + tensorflow experiments with knowledge distillation on EMNIST dataset

pocket-tensor. Run Keras models from a C++ application on embedded devices

keras-shufflenet. ShuffleNet Implementation using Keras Functional Framework 2.0

tensorflow-eager-tutorials. Simple tutorials for building neural networks with TensorFlow Eager mode.

CLR. Jupyter Notebook

gradient-checkpointing. Make huge neural nets fit in memory

AdamW-and-SGDW. keras implementation of AdamW from Fixing Weight Decay Regularization in Adam (https://arxiv.org/abs/1711.05101)

tensorflow. An Open Source Machine Learning Framework for Everyone

ssd_keras. A Keras port of Single Shot MultiBox Detector

nn_playground. Experimental keras implementation of novel neural network structures