awesome-repositories.com
Blog
MCP
awesome-repositories.com

Descubre los mejores repositorios open-source con nuestra búsqueda potenciada por IA.

ExplorarBúsquedas curadasAlternativas open-sourceSoftware autohospedableBlogMapa del sitio
ProyectoServidor MCPAcerca deCómo clasificamosPrensa
Aviso legalPrivacidadTérminos
© 2026 Bringes Technology SRL·VAT RO45896025·hello@awesome-repositories.com
·

4 repositorios

Awesome GitHub RepositoriesHardware-Aware Deployment

Deployment systems that automatically detect host hardware capabilities to select and pull the most optimized model image.

Distinct from Hardware-Agnostic Deployment: Distinct from Hardware-Agnostic Deployment: focuses on active hardware detection and specific image selection rather than generic portability across architectures.

Explore 4 awesome GitHub repositories matching artificial intelligence & ml · Hardware-Aware Deployment. Refine with filters or upvote what's useful.

Awesome Hardware-Aware Deployment GitHub Repositories

Encuentra los mejores repositorios con IA.Buscaremos los repositorios que mejor coincidan usando IA.
  • paddlepaddle/paddle-liteAvatar de PaddlePaddle

    PaddlePaddle/Paddle-Lite

    7,260Ver en GitHub↗

    Paddle-Lite is a deep learning inference engine and edge computing runtime designed to execute trained models on mobile and edge devices. It provides a hardware-accelerated inference framework and a decoupled runtime with a minimal binary footprint to operate in resource-constrained environments without third-party dependencies. The project includes a model quantization tool for reducing precision and size via static and dynamic quantization, as well as a computation graph optimizer. These tools reduce latency and memory usage by fusing operators and pruning the model intermediate representat

    Employs hardware-aware deployment to execute deep learning models across diverse CPU, GPU, and NPU backends.

    C++armbaidudeep-learning
    Ver en GitHub↗7,260
  • pytorch/serveAvatar de pytorch

    pytorch/serve

    4,354Ver en GitHub↗

    Este proyecto es un framework de servicio de modelos de PyTorch diseñado para desplegar y escalar modelos de machine learning en producción a través de endpoints de red escalables. Funciona como un servidor de inferencia de alto rendimiento, optimizador y gestor del ciclo de vida del modelo que maneja la carga de modelos, el procesamiento por lotes (batching) de solicitudes y la aceleración por hardware. El sistema se distingue por sus capacidades avanzadas de orquestación y optimización, como el encadenamiento de múltiples modelos en flujos de trabajo secuenciales mediante grafos de ejecución y el uso de procesamiento por lotes dinámico para mejorar el rendimiento y la latencia. Proporciona soporte especializado para IA generativa y modelos de lenguaje de gran tamaño (LLM) mediante procesamiento por lotes continuo y paralelismo de tensores. Las áreas de capacidad incluyen la gestión de recursos de GPU en hardware diverso como NVIDIA, AMD y Apple Silicon, así como una gestión integral del ciclo de vida del modelo para registro, versionado y escalado de trabajadores. También integra herramientas de observabilidad para rastrear la salud del sistema y el rendimiento del modelo mediante métricas compatibles con Prometheus. El servidor se gestiona a través de una interfaz de línea de comandos utilizada para el control del ciclo de vida y la configuración de parámetros de tiempo de ejecución.

    Deploys pipeline components across CPUs and GPUs to optimize operational costs based on hardware capabilities.

    Java
    Ver en GitHub↗4,354
  • containers/ramalamaAvatar de containers

    containers/ramalama

    2,605Ver en GitHub↗

    Ramalama is a containerized runtime and management tool for large language models. It functions as an OCI AI model manager and registry client, allowing users to package, distribute, and execute AI models as standardized container images. The project differentiates itself by using OCI-compliant distribution for models and retrieval augmented generation assets, enabling the packaging of vector databases into immutable container images. It features hardware-aware image selection that automatically detects GPU or CPU capabilities to pull the most optimized image for the host environment. The sy

    Implements hardware-aware image selection that automatically detects GPU or CPU capabilities to pull the most optimized model image for the host.

    Pythonaicontainerscuda
    Ver en GitHub↗2,605
  • rlinf/rlinfAvatar de RLinf

    RLinf/RLinf

    2,502Ver en GitHub↗

    RLinf is a distributed reinforcement learning orchestrator and embodied AI training framework. It provides the infrastructure to train vision-language-action models and robotic policies using a combination of reinforcement learning and supervised fine-tuning. The system is designed for scaling workloads across GPU clusters, managing the placement of actors, rollout workers, and environment components. It features a specialized robotics data collection pipeline for gathering teleoperated demonstrations and simulation trajectories into standardized replay buffers, alongside a hardware interface

    Launches actors on specific physical nodes and GPUs using custom environment variables and affinity scheduling.

    Pythonagentic-aiembodied-aireinforcement-learning
    Ver en GitHub↗2,502
  1. Home
  2. Artificial Intelligence & ML
  3. Model Optimization
  4. Inference & Deployment
  5. Model Deployment Toolkits
  6. Hardware-Aware Deployment