awesome-repositories.com
Blog
awesome-repositories.com

Descoperă cele mai bune repository-uri open source cu căutare AI.

ExploreazăCăutări recomandateAlternative open-sourceSoftware self-hostedBlogHartă site
ProiectDespreCum realizăm clasamentulPresăServer MCP
LegalConfidențialitateTermeni
© 2026 Bringes Technology SRL·VAT RO45896025·hello@awesome-repositories.com
·

20 repository-uri

Awesome GitHub RepositoriesGraph-Based Workflow Models

Architectural patterns for modeling processes as directed graphs to manage state and sequencing.

Distinguishing note: Focuses on graph-based orchestration rather than linear workflow engines.

Explore 20 awesome GitHub repositories matching software engineering & architecture · Graph-Based Workflow Models. Refine with filters or upvote what's useful.

Awesome Graph-Based Workflow Models GitHub Repositories

Găsește cele mai bune repo-uri cu AI.Vom căuta cele mai potrivite repository-uri folosind AI.
  • wshobson/agentsAvatar wshobson

    wshobson/agents

    36,830Vezi pe GitHub↗

    This project is an automated trading and agentic workflow platform designed to orchestrate complex financial tasks through state-based graphs. It provides a comprehensive framework for building, deploying, and managing autonomous agents that execute multi-step analytical processes, monitor real-time market conditions, and perform high-speed trade execution. The platform distinguishes itself through a robust agentic plugin ecosystem that integrates directly with popular AI-powered development environments and command-line interfaces. It features a specialized financial analysis engine capable

    Models complex multi-step processes as directed graphs to manage agent transitions, tool sequencing, and state persistence.

    Pythonagentsanthropicanthropic-claude
    Vezi pe GitHub↗36,830
  • n8n-io/self-hosted-ai-starter-kitAvatar n8n-io

    n8n-io/self-hosted-ai-starter-kit

    14,997Vezi pe GitHub↗

    This project provides a dockerized AI workflow stack and orchestration templates for deploying a self-hosted AI environment. It establishes a localized infrastructure for building autonomous agents and model chains that process private data on-premises without external cloud dependencies. The environment is designed to support autonomous agent development, allowing models to dynamically select tools, execute shell commands, and interact with local file systems. It includes integrated vector database support to enable retrieval augmented generation and private document analysis. The stack cov

    Links discrete logic blocks and AI models into a directed graph to execute complex multi-step tasks.

    aiai-agentslow-code
    Vezi pe GitHub↗14,997
  • retejs/reteAvatar retejs

    retejs/rete

    12,077Vezi pe GitHub↗

    Rete is a framework for building interactive, node-based visual interfaces and dataflow programming environments. It provides a core engine that processes directed graphs, allowing developers to define modular logic where nodes represent operations and connections represent the flow of data or control. By decoupling the graph logic from the user interface, the framework enables the creation of custom visual editors that can be integrated into various frontend component libraries. The project distinguishes itself through a highly extensible, signal-driven architecture that supports complex req

    Constructs and traverses hierarchical node structures to visualize, manipulate, and execute programmatic logic or system architectures.

    TypeScriptdataflow-programmingflow-based-programminggraph-editor
    Vezi pe GitHub↗12,077
  • nvidia/triton-inference-serverAvatar NVIDIA

    NVIDIA/triton-inference-server

    10,756Vezi pe GitHub↗

    Triton Inference Server is a high-performance AI model inference server and multi-framework model runtime designed for deploying machine learning models across cloud, data center, and embedded edge infrastructure. It serves as an execution engine that allows for the concurrent running of models from various frameworks to optimize hardware utilization. The project features a dynamic batching inference engine that groups individual requests into larger batches to increase total processing throughput. It also provides a model ensemble pipeline, which enables the chaining of multiple models toget

    Provides ensemble pipelines that chain multiple models together into complex data processing sequences.

    Python
    Vezi pe GitHub↗10,756
  • awslabs/autogluonAvatar awslabs

    awslabs/autogluon

    10,481Vezi pe GitHub↗

    AutoGluon is an automated machine learning framework designed to optimize model selection and hyperparameter tuning across tabular, text, image, and time series data. It functions as an ensemble learning library and a tabular data prediction engine, aiming to build high-accuracy predictive models without manual algorithm selection. The framework integrates multimodal machine learning pipelines that combine disparate data types into a single representation using specialized encoders. It also includes a probabilistic time series forecaster that fits multiple statistical and deep learning models

    Orchestrates model pipelines that chain preprocessing and model selection into a consistent workflow.

    Python
    Vezi pe GitHub↗10,481
  • openvinotoolkit/openvinoAvatar openvinotoolkit

    openvinotoolkit/openvino

    10,414Vezi pe GitHub↗

    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

    Orchestrates a sequence of models as a directed acyclic graph to process a single request.

    C++aicomputer-visiondeep-learning
    Vezi pe GitHub↗10,414
  • bentoml/bentomlAvatar bentoml

    bentoml/BentoML

    8,456Vezi pe GitHub↗

    BentoML is a machine learning model serving framework and GPU-accelerated inference server designed to package, deploy, and scale AI models as production-ready REST APIs. It functions as an AI model lifecycle manager and an inference graph orchestrator, enabling the chaining of multiple models and custom logic into complex pipelines for advanced task sequences. The framework distinguishes itself through a dynamic batching engine that optimizes GPU throughput and an artifact-based packaging system that bundles model weights and dependencies into immutable archives for consistent deployment. It

    Enables the chaining of multiple models and custom logic into complex inference graphs for advanced task sequences.

    Pythonai-inferencedeep-learninggenerative-ai
    Vezi pe GitHub↗8,456
  • microsoft/agent-frameworkAvatar microsoft

    microsoft/agent-framework

    7,277Vezi pe GitHub↗

    The agent-framework is an LLM agent orchestration framework and multi-agent workflow engine designed for building autonomous AI agents. It provides a tool integration layer for binding external functions, APIs, and sandboxed code as executable tools for language models. The framework distinguishes itself through a graph-based system for designing sequential and parallel task flows, featuring state management and checkpointing for long-running processes. It implements comprehensive conversational state management and an observability suite that uses telemetry to trace execution flows and monit

    Uses graph architectures to model complex task sequences supporting parallel processing and conditional routing.

    Pythonagent-frameworkagentic-aiagents
    Vezi pe GitHub↗7,277
  • datatalksclub/llm-zoomcampAvatar DataTalksClub

    DataTalksClub/llm-zoomcamp

    6,529Vezi pe GitHub↗

    llm-zoomcamp is a comprehensive educational program and course for building real-life AI systems using large language models. It serves as a structured curriculum and implementation guide for developing AI applications and retrieval techniques. The project provides instructional material on building retrieval augmented generation pipelines to ground model responses in custom knowledge bases. It includes training on vector database implementation, semantic search, and the use of function calling to create autonomous agentic workflows. The curriculum covers a broad range of system development

    Instructional material on chaining multiple AI models in a sequence where outputs flow as inputs to subsequent steps.

    Jupyter Notebook
    Vezi pe GitHub↗6,529
  • kubeflow/kfservingAvatar kubeflow

    kubeflow/kfserving

    5,576Vezi pe GitHub↗

    KServe is an open platform for deploying and serving generative and predictive AI models on Kubernetes. It defines inference services as custom resources with declarative YAML specifications, enabling a Kubernetes-native approach to model deployment and lifecycle management. The platform leverages Knative-based serverless scaling for automatic scale-to-zero and revision management, and supports a pluggable serving runtime architecture that maps model formats to containerized execution environments. KServe distinguishes itself through model-aware autoscaling that scales replicas based on token

    Coordinates inference requests across chains of models with conditional logic and branching.

    Go
    Vezi pe GitHub↗5,576
  • kserve/kserveAvatar kserve

    kserve/kserve

    5,576Vezi pe GitHub↗

    KServe is a Kubernetes-native platform for deploying and serving machine learning models as scalable inference services. It supports both generative AI models, including large language models, and traditional predictive models from frameworks such as TensorFlow, PyTorch, Scikit-Learn, XGBoost, and ONNX. The platform manages the full lifecycle of model deployments, including revision tracking, canary rollouts, A/B testing, and automatic rollbacks, and provides serverless scale-to-zero capabilities for cost-efficient resource management. KServe distinguishes itself through a standardized infere

    Routes inference requests through a directed graph of multiple models, chaining outputs as inputs to downstream steps.

    Go
    Vezi pe GitHub↗5,576
  • seldonio/seldon-coreAvatar SeldonIO

    SeldonIO/seldon-core

    4,752Vezi pe GitHub↗

    Seldon Core este un server de modele de machine learning bazat pe Kubernetes și un framework de inferență MLOps. Funcționează ca un motor de servire multi-model și orchestrator de pipeline-uri, împachetând modelele ca microservicii scalabile care sunt expuse prin API-uri standardizate REST și gRPC. Proiectul se distinge prin pipeline-uri de inferență bazate pe grafuri care înlănțuie modele și transformatoare de date în fluxuri de lucru secvențiale. Optimizează utilizarea hardware-ului prin servire partajată multi-model și strategii de overcommit dinamic al memoriei, susținând în același timp experimentarea în producție prin rutarea ponderată a traficului, testarea A/B și deployment-uri de tip shadow. Framework-ul acoperă o gamă largă de capabilități MLOps, inclusiv autoscaling bazat pe cerere, procesarea asincronă a cererilor prin message bus-uri și monitorizarea completă pentru data drift, valori aberante (outliers) și explicabilitatea predicțiilor. Oferă, de asemenea, gestionarea infrastructurii pentru configurarea runtime-ului modelelor și comunicare securizată folosind criptare TLS pe planurile de control și de date.

    Chains multiple AI models in a directed graph where outputs flow as inputs to subsequent models.

    Goaiopsdeploymentkubernetes
    Vezi pe GitHub↗4,752
  • initialcommit-com/git-simAvatar initialcommit-com

    initialcommit-com/git-sim

    4,663Vezi pe GitHub↗

    git-sim is a suite of tools designed to generate branded animations, visual commit graphs, and technical media for documenting version control workflows. It functions as a git animation generator and command visualizer that simulates how specific operations change repository states and commit graphs. The system models git operations on dummy repositories to predict outcomes and produces videos showing state transitions. These animations include customizable styles, speeds, and branded sequences with custom logos and text in the intro and outro. The project covers the generation of technical

    Represents repository history as a directed graph to calculate state transitions after simulating Git commands.

    Pythongitgitpythonmanim
    Vezi pe GitHub↗4,663
  • pytorch/serveAvatar pytorch

    pytorch/serve

    4,354Vezi pe GitHub↗

    Acest proiect este un framework de servire a modelelor PyTorch conceput pentru a deploya și scala modele de machine learning în producție prin endpoint-uri de rețea scalabile. Funcționează ca un server de inferență de înaltă performanță, optimizator și manager al ciclului de viață al modelelor care gestionează încărcarea modelelor, batching-ul cererilor și accelerarea hardware. Sistemul se distinge prin capabilități avansate de orchestrare și optimizare, cum ar fi înlănțuirea mai multor modele în fluxuri de lucru secvențiale folosind grafuri de execuție și utilizarea batching-ului dinamic pentru a îmbunătăți throughput-ul și latența. Oferă suport specializat pentru AI generativ și modele de limbaj mari (LLM) prin batching continuu și paralelism tensorial. Ariile largi de capabilități includ gestionarea resurselor GPU pe diverse hardware-uri precum NVIDIA, AMD și Apple Silicon, precum și gestionarea cuprinzătoare a ciclului de viață al modelelor pentru înregistrare, versionare și scalarea worker-ilor. De asemenea, integrează instrumente de observabilitate pentru urmărirea stării sistemului și a performanței modelului prin metrici compatibile cu Prometheus. Serverul este gestionat printr-o interfață de linie de comandă utilizată pentru controlul ciclului de viață și configurarea parametrilor de runtime.

    Chains multiple models into sequential workflows and manages the data flow between them in a production graph.

    Java
    Vezi pe GitHub↗4,354
  • huggingface/diffusion-models-classAvatar huggingface

    huggingface/diffusion-models-class

    4,331Vezi pe GitHub↗

    Acest proiect este un curs educațional și o colecție de materiale de antrenament axate pe modele generative de difuzie. Oferă un curriculum și ghiduri practice pentru antrenarea, fine-tuning-ul și implementarea modelelor capabile să sintetizeze imagini, audio și video. Materialul acoperă strategii specifice de implementare, inclusiv sinteza bazată pe zgomot, rafinarea iterativă și compresia spațiului latent. Oferă instrucțiuni despre ghidarea output-urilor generative prin sinteză condiționată și optimizarea aderenței la prompt, precum și tehnici pentru inpainting-ul imaginilor și editarea bazată pe text. Proiectul include conținut despre optimizarea și dezvoltarea modelelor, acoperind fine-tuning-ul conceptelor și reducerea pașilor de inferență. De asemenea, oferă fluxuri de lucru pentru producerea de media sintetică, cum ar fi generarea secvențelor video și convertirea spectrogramelor vizuale în audio. Implementarea practică este oferită prin exemple de cod PyTorch și tutoriale pentru publicarea ponderilor modelelor și a metadatelor pe Hugging Face Hub.

    Implements specialized workflows to orchestrate the process of noise addition and removal for synthetic media.

    Jupyter Notebook
    Vezi pe GitHub↗4,331
  • xerrors/yuxi-knowAvatar xerrors

    xerrors/Yuxi-Know

    4,354Vezi pe GitHub↗

    Yuxi-Know is an LLM agent orchestration platform that coordinates multiple AI agents through graph-based workflows to decompose and execute complex reasoning tasks. It functions as a multi-tenant AI workspace with an agentic chat interface, combining retrieval-augmented generation with knowledge graph management for enterprise document processing and retrieval. The platform distinguishes itself through graph-based agent orchestration, where directed acyclic graphs define execution dependencies between reasoning steps, enabling parallel or sequential task decomposition. It provides multi-tenan

    An engine that decomposes complex tasks into parallel or sequential steps using graph-based agent coordination.

    Pythondockerfastapikbqa
    Vezi pe GitHub↗4,354
  • orchest/orchestAvatar orchest

    orchest/orchest

    4,138Vezi pe GitHub↗

    Orchest este un orchestrator de pipeline-uri de date și un manager de fluxuri de lucru containerizat. Oferă o platformă pentru proiectarea, programarea și executarea secvențelor complexe de procesare a datelor printr-o combinație de interfață grafică și scripting. Platforma se distinge prin utilizarea containerelor pentru a gestiona dependențele software, asigurând o execuție consistentă în medii diferite. Dispune de un scheduler de sarcini poliglot capabil să declanșeze joburi scrise în mai multe limbaje de programare și include un sistem de control al versiunilor care urmărește snapshot-urile istorice ale configurațiilor de proiect și ale codului. Sistemul acoperă designul vizual al fluxului de lucru și maparea dependențelor bazată pe grafuri, alături de programarea sarcinilor declanșate de timp pentru execuție recurentă sau imediată. De asemenea, suportă implementarea serviciilor de fundal persistente care rămân active pe durata rulării unui pipeline.

    Implements graph-based workflow models to map task dependencies and execution sequences visually.

    TypeScriptairflowclouddag
    Vezi pe GitHub↗4,138
  • comfyanonymous/comfyui_examplesAvatar comfyanonymous

    comfyanonymous/ComfyUI_examples

    3,918Vezi pe GitHub↗

    This repository is a collection of node-based pipeline configurations, examples, and templates for generating AI media. It provides a workflow library and a curated gallery of blueprints designed for creating images, videos, and 3D assets using diffusion models. The project specifically offers a set of pre-configured node graphs for implementing advanced image generation and refinement techniques, with a focus on Stable Diffusion workflows. These examples demonstrate how to interconnect processing nodes to define complex generative logic without writing code. The available templates cover a

    Embeds node graphs directly into generated images to allow full recreation of the generation process.

    HTML
    Vezi pe GitHub↗3,918
  • lazyagi/lazyllmAvatar LazyAGI

    LazyAGI/LazyLLM

    3,842Vezi pe GitHub↗

    LazyLLM is a multi-agent framework and orchestration engine designed for building complex AI applications. It provides a system for chaining large language models into sequential or parallel pipelines, utilizing a tool registry to convert standard functions into discoverable tools that models can invoke via reasoning. The project features an application deployment kit that enables hosting model workflows as web services with integrated chat interfaces and API gateways. It includes an infrastructure abstraction layer that allows users to switch between bare-metal servers, clusters, and public

    Orchestrates the chaining of multiple AI models and tools into directed graphs with sequential or parallel flows.

    Pythonagentsai-agentdata
    Vezi pe GitHub↗3,842
  • paddlepaddle/servingAvatar PaddlePaddle

    PaddlePaddle/Serving

    921Vezi pe GitHub↗

    Serving este un framework de înaltă performanță conceput pentru deployment-ul și scalarea modelelor de machine learning ca servicii de producție. Funcționează ca un motor de inferență distribuit care permite execuția fluxurilor de lucru complexe de procesare a datelor prin înlănțuirea mai multor modele în grafuri aciclice direcționate. Platforma se distinge prin capacitatea sa de a gestiona întregul ciclu de viață al modelului de producție, permițând versionarea hot-swappable care actualizează serviciile fără downtime. Suportă scalarea orizontală prin sharding-ul distribuit al modelelor și optimizează regăsirea datelor multidimensionale prin structuri specializate de căutare a parametrilor rari (sparse). Sistemul oferă o suită cuprinzătoare de capabilități pentru mediile de producție, inclusiv execuția inferenței accelerată hardware, interfețe RPC (remote procedure call) multi-limbaj și monitorizarea integrată a serviciilor. De asemenea, încorporează funcții de securitate, cum ar fi autentificarea cererilor și canale de comunicare criptate pentru a proteja deployment-urile modelelor.

    Chains multiple machine learning models into sequential workflows to process complex data tasks with high throughput.

    C++dagdeep-learningdocker
    Vezi pe GitHub↗921
  1. Home
  2. Software Engineering & Architecture
  3. Graph-Based Workflow Models

Explorează sub-etichetele

  • Metadata EmbeddingTechniques for storing graph-based workflow definitions directly within the output media assets. **Distinct from Graph-Based Workflow Models:** Distinct from Graph-Based Workflow Models: focuses on the serialization and storage of the graph into file metadata rather than the structural model of the graph.
  • Model Pipeline Orchestration1 sub-tagSystems for chaining multiple AI models in a directed graph where outputs flow as inputs to subsequent models. **Distinct from Graph-Based Workflow Models:** Specifically for AI model chaining rather than general business process or state-machine workflows.
  • Version Control Graph ModelingModeling repository history and state transitions as directed graphs. **Distinct from Graph-Based Workflow Models:** Specifically applies graph-based workflow modeling to Git commit histories rather than general business processes.