Awesome GitHub RepositoriesIntermediate Output Caching

Sglang is a high-performance inference engine and serving system designed for large language and multimodal models. It provides a programmable interface for orchestrating complex generation workflows, enabling developers to coordinate multi-turn dialogues, tool invocations, and reasoning chains through a domain-specific language. The platform is built to support production-scale deployments, offering an OpenAI-compatible API that allows for integration with existing application ecosystems. The system distinguishes itself through a disaggregated architecture that separates compute-intensive pr

This project is a machine learning array framework and tensor computation library designed for high-performance numerical computing. It provides a comprehensive suite of tools for constructing and training neural networks, featuring an automatic differentiation engine that facilitates gradient-based optimization and complex mathematical modeling. The library distinguishes itself through a unified memory architecture that allows data to be shared across CPU and GPU devices without explicit copies, significantly reducing data movement overhead. Its execution model relies on a lazy evaluation en

This project is a comprehensive framework for building AI-powered applications, providing a unified toolkit for orchestrating language models, autonomous agents, and interactive user interfaces. It serves as a central library for managing the entire lifecycle of AI interactions, from initial prompt generation and model provider abstraction to complex, multi-step reasoning and tool execution. The framework distinguishes itself through its deep integration with frontend development, specifically by enabling generative user interfaces that render dynamic components directly from model outputs. I

Datasets is a library designed for the management, processing, and sharing of large-scale data collections for machine learning workflows. It functions as both a data processing framework and a versioning platform, providing tools to organize, filter, and transform massive datasets while ensuring reproducibility across research and development teams. The library distinguishes itself by enabling the handling of datasets that exceed available system memory. It utilizes memory-mapped file access, disk-based caching, and lazy iterative streaming to maintain performance when working with large-sca

FramePack is a neural video synthesis engine and generation framework designed to produce long, temporally consistent video sequences. It functions as a diffusion model optimizer, providing a suite of techniques to manage the computational demands of high-parameter video models while maintaining visual stability during extended generation tasks. The system distinguishes itself through a hierarchical approach to frame prediction, which plans distant anchor frames before filling in intermediate content to prevent cumulative temporal drift. By utilizing constant-length context compression and to

StreamDiffusion is an interactive generative AI framework and inference engine designed for the low-latency delivery of image and video streams. It provides a real-time Stable Diffusion pipeline for text-to-image and image-to-image generation, enabling the creation of continuous generative image streams with minimized computational delay. The framework optimizes throughput using a pre-computed cache engine and residual-based guidance approximation to reduce the number of required model passes. It further manages GPU load through similarity-based frame skipping, which avoids redundant computat

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

LMCache is a distributed key-value cache manager and tiering system designed to accelerate large language model inference. It functions as a tiered storage layer that offloads tensors from GPU memory to CPU RAM, local disks, or remote object stores, enabling the reuse of cached prefixes across different inference sessions and serving engines. The system differentiates itself through a disaggregated prefill-decode model, which separates prompt processing from token generation by transferring caches between distributed compute nodes. It utilizes peer-to-peer orchestration to share and retrieve

Data-Juicer is an open-source framework for cleaning, filtering, deduplicating, and transforming multimodal datasets to prepare them for training large language and vision models. It functions as a distributed data pipeline engine that runs processing jobs across Ray clusters, handling billions of samples with automatic operator fusion and adaptive parallelism. The framework provides a library of operators that leverage large language models for semantic extraction, filtering, and data synthesis within processing pipelines. The project distinguishes itself through a YAML-based data recipe sys

SparkInternals est une référence technique et un guide d'architecture détaillant la conception interne et l'implémentation du moteur de calcul distribué Apache Spark. Il sert d'étude sur l'analyse des moteurs de big data, en se concentrant sur la gestion de l'exécution en cluster et l'interaction entre les nœuds drivers, les exécuteurs et les workers. Le projet fournit une décomposition détaillée de la manière dont les plans logiques sont convertis en étapes d'exécution physiques. Il analyse spécifiquement la mécanique des opérations de shuffle, la gestion de la mémoire et la coordination de la planification des jobs distribués. La documentation couvre un large éventail de capacités de calcul distribué, incluant la planification de l'exécution des requêtes, la gestion des dépendances de données et les stratégies de mise en cache en mémoire. Elle examine également la distribution des tâches, l'exécution parallèle et les processus utilisés pour la reprise sur erreur et la persistance des données.

YouDub-webui is a multilingual video translator and AI dubbing pipeline manager featuring a web interface for automating video translation, audio dubbing, and subtitle burning. It utilizes a GPU-accelerated media processor to speed up audio transcription and video rendering tasks. The system implements a stage-based pipeline that converts original speech into new languages while preserving background audio through audio track mixing. It supports multiple localization workflows, including automated translation and subtitle-driven dubbing using SRT files to bypass automatic transcription phases

pybroker is a Python algorithmic trading framework and quantitative technical analysis library designed for developing, testing, and optimizing trading strategies using historical market data. It functions as a trading strategy backtester and a financial performance evaluator, providing a structured environment to simulate trading rules and analyze their statistical reliability. The framework distinguishes itself through a market data integration layer that handles the fetching and caching of historical price data from external providers. It incorporates an event-driven backtesting engine and

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