5 dépôts
Configuration parameters that enable optimized execution paths for production workloads.
Explore 5 awesome GitHub repositories matching artificial intelligence & ml · High-Performance Inference Modes. Refine with filters or upvote what's useful.
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
Activates optimized execution paths through specific configuration parameters to boost performance in production environments.
This project is a distributed training infrastructure designed for aligning large language models through reinforcement learning. It functions as an end-to-end engine for complex alignment tasks, including proximal policy optimization, direct preference optimization, and iterative self-play. By providing a unified framework for multi-turn interactions and tool-use scenarios, it enables the development of models capable of reasoning and external environment engagement. The framework distinguishes itself through a decoupled architecture that separates model training from sample generation. This
Accelerates the rollout phase of reinforcement learning using optimized inference engines for efficient sample generation.
This project is a comprehensive toolkit designed for the full lifecycle management of large language and multimodal models. It functions as a unified orchestrator that handles the entire development process, ranging from dataset preparation and supervised fine-tuning to advanced reinforcement learning alignment and production-ready inference deployment. The platform distinguishes itself through a specialized reinforcement learning library that supports complex optimization algorithms, including group relative policy optimization and leave-one-out techniques, to improve model instruction-follo
Serves fine-tuned models using optimized kernels and quantization for efficient production access.
PaddleX is a PaddlePaddle-based framework for building, deploying, and fine-tuning AI model pipelines, with pre-built support for computer vision, OCR, document analysis, and time series tasks. It offers a toolkit of ready-to-use pipelines for image classification, object detection, segmentation, and pose estimation, alongside an end-to-end OCR document analysis pipeline that extracts text, tables, formulas, and layout information. The platform also includes a dedicated time series forecasting pipeline for analyzing historical data to detect anomalies, classify patterns, and predict future val
Ships a high-performance inference plugin that automatically selects the optimal backend and configuration for model predictions.
MindSpore est un framework de deep learning conçu pour construire et entraîner des réseaux de neurones dans des environnements cloud, edge et mobiles. Il fonctionne comme un système d'entraînement distribué et une boîte à outils IA accélérée par le matériel, capable d'exécuter des charges de travail sur CPU, GPU et processeurs IA spécialisés. Le projet inclut un moteur de différenciation automatique qui calcule les gradients par transformation de source et compilation statique. Il permet l'entraînement distribué de modèles en répartissant les charges de travail sur le matériel en utilisant le parallélisme de données et de modèles. Le framework couvre le déploiement IA multiplateforme et l'inférence de modèles, utilisant une boîte à outils haute performance pour accélérer l'exécution et le service. Il fournit une accélération spécialisée pour le matériel Ascend et prend en charge le mappage d'opérateurs agnostique au matériel pour les backends de périphériques hétérogènes. L'environnement peut être installé via un gestionnaire de paquets ou compilé à partir de la source sur des systèmes Linux pour des environnements de processeurs IA standard ou spécialisés.
Includes a toolkit that optimizes model execution via graph fusion and quantization to reduce inference latency.