Open-source alternatives to Plaidml

Deep Java Library is a Java deep learning framework and JVM model inference engine. It provides a high-level API for building and deploying deep learning models within the Java ecosystem, acting as a cross-platform runtime for executing models across CPUs, GPUs, and mobile devices. The library is engine-agnostic, allowing users to switch between different deep learning engines such as PyTorch, TensorFlow, and MXNet while maintaining a single unified API. This enables the deployment of the same model across different backends without changing the application code. The framework supports the f

Llamafile is a machine learning model runner and packager that enables local inference by bundling model weights and runtime environments into a single, self-contained executable. It functions as a cross-platform engine, allowing users to execute large language models and perform speech-to-text tasks directly on their own hardware without requiring external software dependencies or complex installations. The project distinguishes itself by utilizing a specialized binary format that allows the same executable to run natively across multiple operating systems and hardware architectures. It auto

Apache MXNet is a deep learning framework and distributed machine learning library designed for training and deploying neural networks across distributed systems, mobile devices, and hardware accelerators. It functions as a cross-platform runtime and a dynamic dataflow scheduler that optimizes neural network execution. The framework provides a multi-language API, enabling the development of machine learning models using Python, R, Julia, Scala, Go, and JavaScript. It supports high-performance model training and the scaling of workloads across multiple GPUs and machines. The system covers cap

This project is a deep learning framework designed for constructing, training, and deploying neural networks across diverse hardware environments. It functions as a high-performance tensor computation library that provides both imperative and symbolic programming interfaces, allowing developers to balance flexible, step-by-step model building with the efficiency of compiled computation graphs. The framework distinguishes itself through a hybrid execution engine that integrates declarative graph compilation with imperative runtime logic. It supports scalable, distributed training across multip

Ivy is a machine learning framework transpiler and model converter designed to ensure deep learning portability. It serves as a tool for migrating source code and models between different deep learning frameworks while maintaining original functionality. The system enables cross-framework model portability by translating model weights, architectures, and source code. It uses abstract syntax tree based transpilation and computational graph tracing to capture execution flows and rewrite high-level logic into target framework code. The project covers model interoperability through weight-layout

MegEngine is a deep learning framework and automatic differentiation engine used for training and deploying neural networks. It functions as a differentiable programming library that enables the creation of mathematical models where operations are differentiable for gradient-based optimization. The project provides a hardware-agnostic tensor runtime and cross-platform model runtime, allowing models to execute across diverse CPU and GPU hardware architectures. It utilizes a dynamic computational graph engine to build execution graphs on the fly, supporting flexible input shapes and complex con

Clash is a functional hardware compiler and hardware description language compiler that translates strongly typed functional programs written in Haskell into synthesizable VHDL, Verilog, and SystemVerilog code. It serves as a development tool allowing digital designers to write hardware descriptions in functional languages and compile them directly to FPGA targets and physical netlists. The compilation infrastructure covers structural hardware compilation, including graph-based code normalization, functional language translation, and netlist generation. It features type-level clock domain saf

TVM is a machine learning compiler framework designed to convert deep learning models from various frameworks into optimized machine code. It functions as a cross-platform deployment engine that transforms high-level model definitions into efficient, hardware-specific binaries for diverse computing architectures. The system utilizes a multi-level compilation pipeline that decouples algorithm logic from hardware implementation through tensor-operator abstractions. It employs a graph-level intermediate representation to perform cross-operator optimizations and memory planning before lowering co

Burn is a deep learning framework designed for building, training, and deploying neural networks using a modular architecture. As a machine learning library built in Rust, it provides a backend-agnostic computational engine that enables the execution of models across diverse hardware, including central processors, graphics processors, and web runtimes. The framework distinguishes itself through a highly portable design that allows developers to maintain a single workflow for both training and inference across heterogeneous environments. It incorporates advanced optimization techniques such as

PaddleOCR is a comprehensive optical character recognition framework designed for detecting and transcribing text from images and documents into structured, machine-readable formats. It provides a modular computer vision pipeline that decouples image preprocessing, text detection, and character recognition into independent, configurable stages. This architecture supports automated document digitization and multilingual text recognition, capable of identifying text in over one hundred languages across diverse environments ranging from scanned documents to industrial scenes. The framework disti

Keras is a high-level deep learning API used to design, build, and train neural networks for tasks such as computer vision, natural language processing, and time series forecasting. It provides a framework for defining model architectures and optimizing weights through a structured interface. The project is defined by a backend-agnostic design that allows the same model code to run across different compute engines. This multi-backend execution enables users to swap underlying engines to optimize for specific hardware or performance requirements. The system supports distributed model training

The nexa-sdk is an on-device AI SDK and multimodal inference engine designed to run large language, vision, and audio models locally on mobile and desktop hardware. It functions as a local LLM runtime and NPU acceleration framework, enabling the execution of generative and discriminative models without reliance on cloud services. The project distinguishes itself through a dedicated NPU acceleration framework that optimizes model execution on Neural Processing Units to reduce latency and power consumption. It employs hardware-agnostic backend routing to dynamically distribute computations acro

OpenBLAS is a high-performance implementation of the Basic Linear Algebra Subprograms standard designed for numerical computing and matrix operations. It serves as a hardware-accelerated numerical library and optimized math kernel library, providing a computational engine for large-scale matrix multiplication and vector operations. The library distinguishes itself through the use of hand-tuned assembly kernels and SIMD instruction mapping, such as AVX and SVE, to maximize floating-point performance on specific CPU architectures. It features a multi-threaded framework that manages parallel exe

This project is a comprehensive toolkit for on-device speech recognition, synthesis, and audio processing, specifically engineered for Apple Silicon. It provides a framework for building real-time, full-duplex voice agents that operate entirely offline, leveraging native hardware acceleration to maintain performance and privacy. By utilizing optimized machine learning models, the library enables local execution of complex audio tasks without reliance on external cloud services. The library distinguishes itself through its specialized focus on local, high-performance voice interaction. It incl

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

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

TensorFlow.js is a JavaScript machine learning library and browser-based runtime used to build, train, and execute models. It functions as a WebGL accelerated tensor engine, providing a foundation for high-performance linear algebra operations and an automatic differentiation framework for computing gradients. The project distinguishes itself through its ability to run machine learning directly in web environments, supporting both client-side inference and browser-based training. It enables the deployment of Python-based models by converting Keras or TensorFlow models into compatible formats

Einops is a tensor manipulation library that provides a framework-agnostic interface for reshaping, Einstein summation, and multi-dimensional array operations. It serves as an abstraction layer that works across NumPy, PyTorch, TensorFlow, and JAX, allowing for tensor transformations without changing the API. The library distinguishes itself through a declarative notation system that uses readable string patterns to describe tensor rearrangements and reductions. This approach includes an extended Einstein summation interface that supports multi-letter axis names and a named dimension mapping

Leela Chess Zero is a deep learning game AI and neural network chess engine that uses search algorithms to determine optimal moves and evaluate game states. It functions as a UCI chess engine, implementing the Universal Chess Interface standard for compatibility with various graphical user interfaces. The system acts as a hardware-accelerated move calculator, leveraging GPU and CPU backends to accelerate neural network inference. It supports the generation and submission of self-play games to training clients to improve the strength of its neural network models. The engine provides capabilit

Claudes C Compiler is a self-contained C compiler toolchain that parses source code, optimizes intermediate representations, and generates machine code without relying on external assemblers, linkers, or toolchain dependencies. It translates C source files directly into standalone executable binaries, object files, or assembly text across multiple hardware architectures. The project provides built-in assembly and linking capabilities, processing architecture-specific assembly text, encoding instructions, combining object files and static archives, resolving symbols, applying relocations, and

PyMC is a Bayesian probabilistic programming framework used for building probabilistic models and performing Bayesian inference. It provides a probabilistic graphical model library for specifying random variables, priors, and likelihood functions, supported by an MCMC sampling engine and variational inference tools to estimate posterior distributions. The framework features a GPU-accelerated inference backend that compiles models into machine code to increase execution speed. It utilizes a backend-agnostic tensor execution model and just-in-time graph compilation to optimize the computation o

whisper.cpp is a C++ implementation of the Whisper speech-to-text model, serving as a lightweight machine learning inference engine and quantized runtime. It provides high-performance automatic speech recognition and real-time audio transcription without requiring a Python environment. The project utilizes model quantization to reduce memory usage and increase inference speed on local hardware. It incorporates hardware acceleration to optimize processing speed across different processors. The system covers audio processing capabilities including voice activity detection, speaker diarization,

Trax is a deep learning framework and hardware-agnostic tensor engine designed for designing and training neural networks. It serves as a research tool providing high-level combinators for composing complex architectures, alongside a dedicated library for building transformer models and a toolkit for reinforcement learning. The framework is distinguished by its support for reversible and sparse transformer architectures, which reduce memory and computational overhead. It enables a single set of model instructions to execute across different hardware backends without changing the underlying co

PaddleFormers is a framework for the training, fine-tuning, and deployment of large language models. It provides a full lifecycle pipeline for executing large-scale model training and applying adaptation methods to align models with specialized tasks. The project focuses on scaling model operations through distributed training and hardware accelerator integration. It employs pipeline parallelism and mixed-precision training to manage memory and increase throughput across multiple hardware devices. The library includes a curated model zoo for serving pre-trained architectures and tools for pr

AITemplate is an ahead-of-time deep learning compiler that translates PyTorch neural networks into standalone C++ source code. It functions as a PyTorch to C++ compiler and a GPU kernel fusion engine, producing self-contained executable binaries that run inference without requiring a Python interpreter or deep learning framework runtime. The project generates optimized CUDA and HIP C++ code specifically for NVIDIA TensorCores and AMD MatrixCores. It focuses on maximizing throughput for half-precision floating-point operations through a system that combines multiple neural network operators in

This is a PyTorch-based computer vision library for detecting 2D and 3D facial landmark coordinates. It functions as a facial landmark detector and reconstruction tool, utilizing deep learning to identify precise geometric points on human faces from image datasets. The library allows for the selection of specific detection backends to balance accuracy and processing speed. It supports the integration of precomputed bounding box files, which enables the system to bypass the initial detection phase and proceed directly to landmark extraction. The toolkit includes capabilities for batch image p

Neural Compressor is a deep learning model compression toolkit and AI inference acceleration engine. It functions as an automated model quantization tool and hardware-aware model compiler designed to reduce the memory footprint of neural networks and decrease execution latency. The project provides specialized frameworks for optimizing large language models, utilizing weight-only quantization and hardware-specific kernels to improve the operational efficiency of generative AI workloads. It maps neural network operators to specialized CPU and GPU vector instructions to accelerate model executi