Awesome GitHub RepositoriesTraining Memory Optimizers

Swin-Transformer is a deep learning framework designed for training and deploying hierarchical vision transformer models. It serves as a research library and toolkit for computer vision tasks, providing the infrastructure to build models that replace standard convolution operations with sliding window self-attention mechanisms. By utilizing a multi-scale feature hierarchy, the framework enables the processing of visual data at varying resolutions and spatial scales. The project distinguishes itself through its implementation of shifted window partitioning, which facilitates global information

Swift is a toolkit for the full-parameter and parameter-efficient fine-tuning of large language and multimodal models. It functions as a multimodal model trainer for text, image, video, and audio data, and includes specialized tools for model compression and reinforcement learning from human feedback. The framework provides an alignment toolkit for optimizing model behavior using preference learning algorithms and reinforcement learning. It integrates parameter-efficient fine-tuning methods to adapt models with minimal memory and compute requirements, alongside utilities for reducing hardware

MNN is a high-performance inference engine and framework designed for on-device machine learning. It provides a comprehensive environment for executing, optimizing, and deploying neural network models directly on mobile and resource-constrained edge devices. The framework distinguishes itself through a robust model optimization toolkit that supports quantization, compression, and structural graph manipulation to minimize memory footprint and maximize execution speed. It features a modular architecture that abstracts hardware-specific backends, allowing models to run efficiently across diverse

Axolotl is a distributed training orchestrator and fine-tuning framework for large language models, multimodal systems, and quantized models. It provides a structured environment for specializing pre-trained models through full parameter updates or low-rank adaptation, as well as aligning model outputs with human expectations via preference tuning pipelines and reward modeling. The system distinguishes itself through a configuration-driven pipeline that manages preprocessing and training workflows via a single file for reproducibility. It implements high-throughput optimizations such as multi

This project is a quantized fine-tuning framework for large language models. It implements a low-rank adaptation library and a four-bit quantizer to reduce the GPU memory requirements needed to train large models. The framework utilizes four-bit quantization and low-rank adapters to enable model training on consumer-grade hardware. It further reduces the memory footprint through double quantization and a paged optimizer that offloads states to system RAM. The system supports distributed training across multiple GPUs to handle larger parameter scales and includes utilities for custom dataset

OpenRLHF is a training framework and alignment library designed for reinforcement learning from human feedback across distributed GPU clusters. It provides tools for aligning large language models and multimodal vision-language models using algorithms such as PPO, GRPO, and DPO. The framework distinguishes itself through a distributed inference engine that overlaps sample rollout with training to increase throughput. It supports scaling to models exceeding 70 billion parameters via parameter sharding and handles long-context sequences through ring-attention sequence parallelism. The project

Sana is a framework for high-resolution image and video synthesis based on a linear diffusion transformer. It provides a toolkit for the training, fine-tuning, and execution of text-to-image and text-to-video models, as well as a video generative world model capable of simulating physical environments with precise spatial control. The project is distinguished by its use of linear complexity layers to handle high resolutions and its support for long-form, minute-length video generation in real time. It implements a two-stage inference paradigm that separates structural generation from visual t

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

bitsandbytes is a deep learning quantization tool and library designed to reduce the memory footprint of large language models. It serves as a GPU memory optimizer and quantization framework, compressing model weights and features to 8-bit and 4-bit precision to enable inference and training on hardware with limited memory. The project provides a framework for low-rank adaptation, allowing the fine-tuning of quantized models by combining 4-bit weights with small trainable matrices. It further distinguishes itself through memory paging, which moves optimizer states between CPU and GPU memory t

هذا المشروع عبارة عن إطار عمل لاكتشاف الكائنات في PyTorch ينفذ بنية Faster R-CNN. يعمل كنموذج رؤية للتنبؤ بصناديق التحديد الدقيقة حول كائنات متعددة داخل الصور وتدفقات الفيديو الحية. تم تحسين النظام للتدريب متعدد الـ GPU لتقليل الوقت المطلوب لتقارب النموذج. يستخدم تصميماً مسرعاً بواسطة GPU للتعامل مع تدريب واستنتاج شبكات الاكتشاف المعقدة. يغطي إطار العمل دورة حياة اكتشاف الكائنات الكاملة، بما في ذلك تدريب الشبكة المخصص والاستنتاج للصور الثابتة وتدفقات الفيديو في الوقت الفعلي. يتضمن قدرات للتحقق من أداء النموذج باستخدام مجموعات بيانات قياسية، بالإضافة إلى تحسينات التدريب مثل التجميع القائم على نسبة العرض إلى الارتفاع وأحمال العمل الموزعة.

This project is a comprehensive educational curriculum and structured learning path covering the full lifecycle of large language models. It provides a guided progression through the theory, architecture, training, and deployment of these models. The curriculum includes specialized guides on transformer architecture, model training tutorials, and frameworks for designing autonomous agents. It also provides dedicated resources for studying model safety and ethics. The material covers a wide range of technical capabilities, including distributed training strategies, parameter-efficient fine-tu

DeepSpeedExamples is a collection of reference implementations and scripts for training, fine-tuning, and executing inference on large-scale AI models using DeepSpeed optimization. It provides a distributed model training guide and practical workflows for adapting large language models through memory-efficient techniques. The repository includes specialized implementations for pipeline parallelism to handle models exceeding single GPU memory and a suite of examples for ZeRO memory optimization to reduce per-device overhead. It also features standardized test suites for benchmarking the throug

Tensorpack هو إطار عمل شبكة عصبية TensorFlow عالي المستوى ومكتبة بحثية مصممة لبناء وتدريب نماذج التعلم العميق. يوفر مجموعة من بنيات الشبكات العصبية القابلة للتكرار للرؤية الحاسوبية، والمهام التوليدية، والتعلم التعزيزي، ومعالجة اللغات الطبيعية. يتميز المشروع بخط معالجة بيانات تعلم عميق متخصص يستخدم Python الخالص لتحميل البيانات المتوازي والبث. ويتضمن منسق تدريب متعدد وحدات GPU لتوزيع أعباء العمل عبر استراتيجيات موازية للبيانات ومجموعة أدوات قابلية تفسير مخصصة لتصور خرائط بروز وتنشيط النموذج. يغطي إطار العمل مجموعة واسعة من القدرات، بما في ذلك خطوط معالجة الرؤية الحاسوبية لاكتشاف الكائنات والتجزئة الدلالية، ونمذجة التسلسل للكلام والنص، وتطوير وكيل التعلم التعزيزي. كما يوفر أدوات تحسين النموذج لتكميم الأوزان والتدريب منخفض البت، إلى جانب مرافق لإعادة إنتاج الأوراق البحثية الأكاديمية وتحويل أوزان نموذج Caffe القديمة.

Tensorpack هو إطار عمل تدريب TensorFlow عالي الأداء ومجموعة أدوات تعلم عميق موزعة. يوفر مجموعة من الأدوات لبناء وتدريب الشبكات العصبية مع التركيز على سرعة التنفيذ ومرونة البنية. يعمل المشروع كجناح لتحسين الشبكات العصبية، حيث يطبق أنماط تنفيذ عالية الكفاءة لتقليل حمل التدريب. ويعمل كخط معالجة بيانات متوازي، باستخدام التوازي التلقائي لزيادة معدل النقل عند معالجة مجموعات البيانات الكبيرة. تغطي مجموعة الأدوات التدريب الموزع عبر وحدات GPU متعددة ومجموعات الحوسبة باستخدام استراتيجيات موازية للبيانات. تشمل قدراتها معالجة مجموعات البيانات واسعة النطاق وتحسين الأداء لزيادة معدل نقل التدريب.

Liger-Kernel is a collection of pre-built fused Triton kernels and patching utilities designed to accelerate large language model training. It provides drop-in kernel replacements for common LLM operations such as RMSNorm, cross-entropy loss, and attention, enabling increased throughput and reduced memory usage while preserving bitwise-exact gradients. The project serves as a toolkit for composing custom model architectures from individual optimized kernels and for patching pre-existing models with minimal code changes. The project distinguishes itself through its ability to perform runtime m

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

Torchtune is a PyTorch-native library for fine-tuning, aligning, and quantizing large language models. It provides a configurable training pipeline orchestrated through YAML recipes, with CLI overrides and component swapping, distributed training via FSDP2, memory optimizations, and parameter-efficient fine-tuning methods like LoRA, DoRA, and QLoRA. The library distinguishes itself through its YAML-driven configuration system that defines all training parameters and instantiates components from config files, with full CLI override capability for any field or component at launch time. It suppo

RF-DETR is a Python library for training and deploying object detection, instance segmentation, and keypoint detection models built on a vision transformer architecture. It provides a unified command-line interface and Python API for the full workflow, from fine-tuning pretrained checkpoints on custom datasets to running inference on images, video files, and live camera streams. The project supports training on datasets in COCO or YOLO format, with automatic format detection and configurable augmentation pipelines. Models can be exported to ONNX, TFLite, or TensorRT for deployment across edge

Composer هو إطار عمل للتدريب الموزع في PyTorch مصمم لتوسيع نطاق النماذج الضخمة عبر مجموعات GPU متعددة العقد. يعمل كمدرب لنماذج اللغات الكبيرة، ومحسن نماذج موزع، ومدير لدورة حياة التدريب. يتميز المشروع كونه مكتبة لتنظيم التعلم العميق (regularization)، حيث يوفر تقنيات تحسين متخصصة مثل Sharpness Aware Minimization وMixUp وCutMix لتحسين تعميم النموذج. كما يميز تدفق التدريب الخاص به باستخدام تسخين طول التسلسل، وتجميد الطبقات التدريجي، وحفظ نقاط التحقق (checkpointing) المجزأة لاستعادة النماذج الضخمة. يغطي إطار العمل مساحة واسعة من القدرات بما في ذلك تنسيق التدريب الموزع، وإدارة الأجهزة ذات الدقة المختلطة، وبث البيانات السحابي. كما يوفر أدوات مراقبة وتشخيص واسعة النطاق لتشخيص ذاكرة GPU، واكتشاف تباعد التدريب، وتتبع الإنتاجية. يتضمن المشروع مشغلاً عبر سطر الأوامر لأتمتة تنفيذ مهام التدريب متعددة الـ GPU عبر العقد.