Awesome GitHub RepositoriesModel Compatibility Layers

This project is a modular research toolkit designed for developing, training, and evaluating deep learning models for object detection, segmentation, and video instance tracking. It provides a flexible training engine that manages complex neural network execution, including distributed training, custom lifecycle hooks, and weight optimization. The framework is built around a hierarchical configuration system that allows users to define architectures, data pipelines, and training hyperparameters through composable, inheritable files. The project distinguishes itself through its highly modular

Paddle is a deep learning framework designed for building, training, and deploying neural networks. It provides a platform for constructing models using tensor-based computations and supports both dynamic and static execution graphs to facilitate research and production workflows. The platform functions as a distributed machine learning system, enabling the scaling of training workloads across multiple nodes and hardware clusters. It includes a comprehensive toolkit for model deployment and optimization, allowing users to convert external model formats, compress trained models for resource-co

ONNX is an open-source standard for machine learning interoperability that provides a unified format for representing neural network models. By defining a common set of operators and a standardized file structure, it enables models to be shared, exported, and executed consistently across different training frameworks and software ecosystems. The project functions as an intermediate representation layer that decouples model development from deployment. It utilizes a language-neutral binary serialization format to store model structures and weights, ensuring that computational graphs remain por

GPT-Neo is an open-source distributed training framework designed for scaling GPT-2 and GPT-3-style language models across multiple devices using mesh-tensorflow for model parallelism. It provides the infrastructure to train transformer-based language models with billions of parameters across distributed computing environments, making large-scale language model research accessible outside of proprietary systems. The framework supports training both autoregressive GPT-style models and masked language models like BERT or RoBERTa, with configurable masking strategies and token handling. It inclu

MergeKit is a toolkit for combining multiple pre-trained large language models into a single entity using algorithmic blending. It provides a specialized system for parameter interpolation and weight extraction to unify model capabilities. The project distinguishes itself through an evolutionary merge optimizer that tunes parameters based on quantitative evaluation metrics. It also features a mixture of experts orchestrator capable of converting dense models into sparse architectures and a tokenizer alignment tool for transplanting embeddings between different models. The toolkit covers a br

Mergekit is a toolkit for combining multiple pretrained large language models into a single model. It functions as an architecture assembler and merging system that transfers capabilities between models using weighted algorithms and layer-wise assembly without requiring additional training. The project provides specialized utilities for extracting low-rank approximations from fine-tuned models to create portable parameter updates. It also includes a framework for converting dense language models into a mixture of experts architecture by constructing gating mechanisms to route inputs to specia

Vespa is a distributed search engine, vector database, and machine learning ranking engine. It serves as an AI search platform designed to handle large-scale document indexing and complex query processing across a cluster of nodes, combining keyword retrieval with high-dimensional embedding storage for semantic similarity search. The platform distinguishes itself by integrating machine learning models directly into the search pipeline to perform real-time inference and ranking. It converts these models into ranking expressions to score and order results based on relevance, while providing a s

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

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

This project is a deep learning model compiler and parser that converts ONNX models into optimized TensorRT engines. It functions as a bridge that maps standardized ONNX operators to vendor-specific kernels to enable high-performance inference on NVIDIA GPUs. The system operates as a GPU inference optimizer, selecting hardware-specific kernels and tuning memory allocation to maximize throughput. It transforms neural network graphs into serialized binary execution plans to reduce runtime overhead. The toolset covers deep learning model deployment and edge AI performance tuning. It includes ca

TransformerLens is a library for mechanistic interpretability research designed to reverse engineer the learned algorithms within large language models. It provides a standardized framework for wrapping diverse transformer architectures, allowing researchers to extract, manipulate, and analyze internal activations and weights through a consistent interface. The project distinguishes itself through a comprehensive system of activation hooks that can capture, patch, and ablate internal tensors during the forward pass. It includes specialized utilities for decomposing fused projections, material