Open-source alternatives to Model Optimizer

Torchtune is a PyTorch-native library for fine-tuning, aligning, and quantizing large language models. It provides a config-driven system for instantiating components, orchestrating distributed training, and managing parameter-efficient fine-tuning with quantization support, all through YAML-based configurations and command-line overrides. The library distinguishes itself through its comprehensive post-training workflow orchestration, combining supervised fine-tuning, preference optimization (DPO, PPO, GRPO), knowledge distillation, and quantization-aware training in a single configurable pip

bitsandbytes is a quantization library for large language models that reduces memory footprints using k-bit quantization. It provides a framework for 4-bit low-rank adaptation, tools for 8-bit model compression, and memory-efficient optimizer extensions for PyTorch. The project enables the training of large models on limited hardware through 4-bit quantization and low-rank adaptation weights. It also facilitates faster inference by compressing models to 8-bit precision using vector-wise quantization. The library covers a range of memory optimization capabilities, including optimizer memory r

PyTorch Lightning is a high-level deep learning framework for PyTorch that automates training loops and removes repetitive engineering boilerplate. It functions as a structured pipeline for managing machine learning experiments, providing a distributed training orchestrator and tools for mixed-precision training. The framework decouples scientific model architecture from the engineering required for infrastructure and scaling. This separation allows the same model code to execute across CPUs, GPUs, or TPUs through a hardware-agnostic execution engine and a centralized trainer that manages the

This project is a vision language model framework and vision-to-text pipeline designed for deploying and optimizing models that process both images and text. It provides an on-device inference engine and a vision language model framework to run quantized models locally on mobile and desktop hardware accelerators. The framework features a model quantization toolkit to reduce weight precision for lower memory footprints and increased execution speed on specialized silicon. It also includes an efficient vision encoder utilizing a hybrid encoding system to compress image tokens, which reduces pro

AISystem is a comprehensive AI full-stack infrastructure project covering the entire pipeline from AI chip architecture to high-level training frameworks. It encompasses the development of AI compiler frameworks, inference engines, and distributed training orchestrators designed to coordinate workloads across a heterogeneous compute stack of CPUs, GPUs, and NPUs. The project focuses on the deep integration of software and hardware, employing software-hardware co-design to align tensor layouts with physical memory structures. It provides specialized capabilities for accelerating Transformer mo

ipex-llm is an acceleration library and inference engine designed to optimize the execution and finetuning of large language models on Intel GPUs and NPUs. It provides a HuggingFace compatible model backend and a dedicated quantization toolkit for converting model weights into low-bit precision formats. The project facilitates distributed inference by splitting large model workloads across multiple accelerators using pipeline and tensor parallelism. It enables the deployment of models on Intel Arc, Flex, and Max GPUs to increase throughput and reduce latency. The library covers a broad range

llm-compressor is a quantization toolkit and post-training library designed to reduce the memory footprint and size of large language models. It provides a framework for compressing models using weight and activation quantization to enable more efficient deployment. The project distinguishes itself through a distributed quantization framework that utilizes data-parallel processing and disk-based weight offloading to handle massive model checkpoints that exceed available system memory. It includes specialized compressors for diverse architectures, including Mixture-of-Experts, Vision-Language,

This project is a comprehensive collection of educational examples and reference implementations for building vision and language models using PyTorch. It serves as a deep learning tutorial covering the end-to-end process of developing neural networks, from initial architecture definition to final production deployment. The repository provides detailed guides on implementing a wide range of domain-specific models, including convolutional neural networks for object detection and segmentation, as well as transformer and recurrent architectures for natural language processing. It emphasizes gene

Darts is a differentiable architecture search framework and library designed to automate the discovery of optimal convolutional and recurrent neural network structures. It serves as a research tool for finding high-performing cell topologies using gradient-based optimization. The framework employs a differentiable cell super-net and weight-sharing mechanisms to identify effective network connectivity. It utilizes second-order approximation to estimate the performance of discrete architectural candidates and converts learned continuous weights into discrete graph structures through genotype-to

OpenVINO is an AI inference engine and model serving platform designed to execute optimized deep learning models across CPUs, GPUs, and NPUs through a unified API. It includes a model optimization toolkit for converting, quantizing, and compressing models from various frameworks, alongside a specialized generative AI runtime for large language models. The project distinguishes itself through a plugin-based hardware acceleration layer that maps neural network operations to vendor-specific drivers. It features advanced execution mechanisms such as continuous batching, speculative decoding, and

PowerInfer is an inference engine and serving framework designed to run large language models on local hardware. It combines a hybrid CPU-GPU offloader, a quantization tool, and a sparse model optimizer to enable the execution of high-parameter models on consumer-grade devices. The system distinguishes itself through neuron-activation-based offloading, using a predictor model to preload frequent neurons into VRAM while keeping rare neurons in system memory. This hybrid execution model balances workloads between the GPU and CPU based on input patterns to optimize memory access and increase tok

pykan is a library for implementing Kolmogorov-Arnold Networks, replacing fixed node activation functions with learnable spline functions located on the network edges. It serves as an interpretable AI framework and symbolic regression tool designed to derive transparent mathematical rules from complex data. The project focuses on converting learned numerical functions into human-readable symbolic expressions through library matching and formula conversion. It utilizes additive-compositional topologies and learnable piecewise polynomial segments to approximate non-linear mappings. The framewo

This project is a PyTorch library for building and training Kolmogorov-Arnold Networks. It implements a neural network architecture that replaces fixed activation functions with learnable spline-based functions on edges, serving as a tool for interpretable machine learning. The implementation utilizes reformulated matrix operations to reduce memory overhead and increase computation speed. It employs L1 regularization to sparsify network weights, which improves the transparency of the model's internal logic and decisions. The framework covers a range of capabilities including grid-based funct

This project is a comprehensive instructional resource and course for building neural networks using PyTorch. It covers the fundamental building blocks of deep learning, including tensor manipulation, automatic differentiation, and the construction of modular neural network components. The repository serves as a technical guide for several specialized domains. It provides implementation details for computer vision tasks such as image classification, object detection, and semantic segmentation, as well as natural language processing workflows involving transformers, recurrent networks, and gen

Caffe is a high-performance deep learning framework and convolutional neural network library designed for training and deploying neural networks. It functions as a GPU-accelerated machine learning engine with a core implemented in C++ to enable high-throughput tensor operations. The project utilizes a declarative configuration system where model architectures and hyperparameters are defined in external text files, separating the network design from the execution code. It includes a model serialization system to export trained weights and topologies into binary files for efficient deployment a

PaddleClas is a toolkit for image classification and recognition built on PaddlePaddle. It provides a suite of tools for training deep learning models and a framework for implementing visual search and retrieval systems. The project includes a computer vision model optimization suite and tools for cross-platform deployment. It enables the export of trained models to servers, mobile devices, and edge hardware to achieve high-performance inference across different programming languages. The toolkit covers model compression and optimization through pruning, quantization, and knowledge distillat

This project is a structured learning curriculum and technical reference for mastering deep learning with TensorFlow. It provides a comprehensive guide for building, training, and deploying neural networks, combining theoretical fundamentals with practical implementation examples. The repository distinguishes itself by covering the end-to-end machine learning workflow, from low-level tensor mathematics and linear algebra to the creation of complex model architectures. It includes specific guidance on developing data pipelines for diverse data types, such as images, text, and time-series seque

PaddleNLP is a development library and toolkit for training, fine-tuning, and deploying large and small language models using the PaddlePaddle framework. It provides a comprehensive suite for the entire natural language processing lifecycle, from model development to high-performance inference. The project features a standardized model zoo for loading and managing pre-trained models and tokenizers through a unified interface. It distinguishes itself with a specialized model compression framework that reduces memory footprints via weight precision conversion and lossless size optimization, alo

NNI is an AutoML toolkit designed to automate machine learning lifecycles. It functions as a hyperparameter optimization framework, a neural architecture search tool, and a model compression suite. The project provides a distributed training orchestrator to manage machine learning workloads across local machines, remote servers, and cloud platforms. It enables the discovery of efficient model structures through reinforcement learning and one-shot optimization methods, while utilizing Bayesian and evolutionary algorithms to automate hyperparameter tuning. Additional capabilities include tools

OpenNMT-py is a PyTorch neural machine translation framework used for training and deploying neural machine translation and large language models. It functions as a distributed model training system, an inference engine, and a toolkit for fine-tuning large language models. The framework distinguishes itself with a dedicated toolkit for adapting large language models through low-rank adaptation, quantization, and instruction tuning. It also includes a neural machine translation server that allows trained models to be hosted and exposed via REST API endpoints. The project covers a broad range

Anomalib is a PyTorch-based library for visual anomaly detection, offering a modular framework, a comprehensive model zoo, and a benchmarking suite designed for industrial defect detection. It provides a wide range of algorithms—including generative, discriminative, teacher-student, and vision-language approaches—that support unsupervised, few-shot, and zero-shot settings. The library enables deployment through model export to ONNX and OpenVINO for edge devices, and includes a no-code web application for training and inference. It also features a command-line interface for orchestrating multi

This project is a comprehensive educational resource and tutorial handbook for building, training, and deploying machine learning models using TensorFlow 2. It serves as a structured learning guide covering core deep learning concepts, including neural network architectures, automatic differentiation, and tensor operations. The handbook provides technical guidance on optimizing execution efficiency through GPU memory management, distributed training, and model quantization. It also includes detailed manuals for constructing high-performance data pipelines and exporting models for production s

Composer is a PyTorch distributed training framework designed for scaling large-scale models across multi-node GPU clusters. It functions as a large language model trainer, a distributed model optimizer, and a training lifecycle manager. The project differentiates itself as a deep learning regularization library, providing specialized optimization techniques such as Sharpness Aware Minimization, MixUp, and CutMix to improve model generalization. It further distinguishes its training flow through the use of sequence length warmup, progressive layer freezing, and sharded-state checkpointing for

This project provides a comprehensive collection of educational resources and technical guides for training, fine-tuning, and deploying machine learning models using PyTorch and Hugging Face. It serves as a practical reference for scaling deep learning workflows, offering structured instructions for managing large-scale architectures across distributed hardware accelerators. The repository distinguishes itself by focusing on the end-to-end lifecycle of large language models, specifically emphasizing containerized deployment and performance optimization. It details workflows for parameter-effi

Distil-Whisper is a compressed automatic speech recognition model designed to convert spoken audio into written text. It uses a transformer-based sequence-to-sequence architecture to provide speech-to-text transcription. The project utilizes knowledge distillation and teacher-student model compression to create a lightweight version of the Whisper model. This approach reduces GPU memory usage and accelerates token prediction while maintaining transcription accuracy. The system supports both short-form audio transcription and long-form processing through the use of sliding windows and chunked

FasterTransformer is a high-performance inference optimization library and distributed runtime designed to accelerate the execution of transformer models. It provides a toolkit for reducing model precision and parallelizing execution across multiple GPUs to increase throughput and reduce latency for large language models. The framework utilizes a C++ backend with custom CUDA kernels to replace generic operations with optimized GPU instructions. It implements tensor and pipeline parallelism to shard model weights and distribute compute operations across multiple devices. The system includes c

PaddleGAN is a generative AI framework and deep learning computer vision library built on the PaddlePaddle framework. It serves as a toolkit for image and video synthesis, providing a collection of generative adversarial network implementations for creating synthetic visual content. The library focuses on advanced synthesis capabilities, including the generation of talking heads through lip motion synchronization and the creation of synthetic videos via motion transfer from driving sequences. It provides tools for domain-to-domain translation, allowing for image style transfer and the transfo