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

Découvrez les meilleurs dépôts open-source grâce à notre recherche par IA.

ExplorerRecherches sélectionnéesAlternatives open sourceLogiciels auto-hébergésBlogPlan du site
ProjetÀ proposNotre méthodologiePresseServeur MCP
Mentions légalesConfidentialitéConditions d'utilisation
© 2026 Bringes Technology SRL·VAT RO45896025·hello@awesome-repositories.com
·
FareedKhan-dev avatar

FareedKhan-dev/all-agentic-architectures

0
View on GitHub↗
3,670 stars·649 forks·Jupyter Notebook·MIT·9 vuesfareedkhan-dev.github.io/all-agentic-architectures↗

All Agentic Architectures

Ce projet est un framework complet pour construire, évaluer et connecter des systèmes d'agents autonomes. Il fournit une bibliothèque de modèles architecturaux standardisés pour implémenter des workflows d'agents complexes, y compris l'orchestration multi-agents, le raisonnement itératif et la gestion de la mémoire. En offrant une interface unifiée pour les fournisseurs de modèles, le framework permet une exécution cohérente des agents à travers différents services d'intelligence artificielle.

Le framework se distingue par une focalisation sur le benchmarking rigoureux et le contrôle déterministe. Il inclut une suite d'outils pour évaluer la performance des agents par rapport à des tâches standardisées et des métriques de qualité, permettant la comparaison de différents modèles de conception. Pour garantir la fiabilité, le système intègre des portes de routage déterministes et des boucles d'auto-correction qui valident les actions des agents et affinent les sorties par rapport aux critères de qualité avant l'exécution externe.

L'architecture prend en charge un large éventail de capacités, y compris l'intégration d'outils pour l'accomplissement de tâches dans le monde réel, la génération augmentée par récupération (RAG) pour des réponses sensibles au contexte, et la gestion modulaire de la mémoire pour maintenir l'information à travers les sessions. Ces composants sont liés par un contrat d'exécution standardisé qui garantit un comportement cohérent indépendamment du modèle sous-jacent ou de la configuration architecturale spécifique.

Le dépôt est structuré comme une collection de Jupyter Notebooks qui démontrent ces modèles et méthodologies de benchmarking.

Features

  • Agentic LLM Frameworks - Provides a comprehensive framework for building and benchmarking autonomous agent architectures with integrated reasoning and memory.
  • Agentic Reasoning Loops - Facilitates self-correction loops where agents critique, verify, and revise their own outputs until the final result meets defined quality standards.
  • AI Agent Architectures - Provides a library of standardized patterns for implementing complex agent workflows and architectures.
  • Autonomous Agent Patterns - Implements standardized architectural patterns for reasoning, retrieval, and memory to construct autonomous agents.
  • External Tool Integrations - Enables agents to interact with external environments by executing tools such as web searches, code interpreters, and browser automation.
  • LLM Provider Interfaces - Provides a unified abstraction layer for connecting autonomous agents to various large language model providers through a consistent execution contract.
  • Execution Contracts - Enforces a uniform execution protocol to ensure consistent behavior and benchmarking across different agent configurations.
  • Memory Storage and Retrieval Systems - Provides structured storage systems for maintaining episodic and semantic context across agent sessions.
  • Deterministic Decision Gates - Prevents reasoning errors by using categorical scoring and programmatic logic to enforce strict boundaries on agent outputs during the decision-making process.
  • Agent Execution Policies - Provides standardized execution policies and protocols for consistent agent instantiation and performance measurement.
  • Agent Evaluation Frameworks - Assesses the reliability of agent responses by using automated judge models to score system effectiveness against predefined quality benchmarks.
  • Multi-Agent Orchestration Frameworks - Coordinates multiple specialized agents through debate and ensemble workflows to solve complex problems.
  • Model Provider Integrations - Enables developers to swap between different artificial intelligence model providers using a unified interface without requiring infrastructure code changes.
  • Action Approval Gates - Implements programmatic gates to validate agent actions against safety constraints before external execution.
  • Autonomous Agents - Supports the development of complex systems where agents manage multi-step processes and long-term memory autonomously.
  • Search-Based Reasoning Strategies - Supports exploring multiple decision trees and reasoning trajectories to select optimal answers through sampling and reward-based search methods.
  • Retrieval-Augmented Generation - Grounds agent responses by integrating adaptive and graph-based search strategies for context-aware retrieval.
  • Retrieval Augmented Generation - Implements graph-based and adaptive search strategies to ground agent responses in verifiable external data.
  • Provider Abstraction Layers - Provides a unified interface layer to translate requests and enable seamless switching between different AI service providers.
  • Multi-Agent Orchestration Systems - Orchestrates multiple specialized agents through collaborative workflows to solve complex problems.
  • Self-Correction Loops - Enables recursive refinement cycles where agents evaluate and correct their own outputs against quality criteria.
  • Tree Search Reasoning Solvers - Explores multiple decision paths using tree-based search algorithms to select optimal reasoning trajectories.
  • Tool Use And Integration - Connects agents to external environments like web browsers and code interpreters for real-world task completion.
  • Agent Memory Management - Maintains interaction context using episodic and semantic structures to ensure continuity across tasks.
  • LLM Safety Enforcers - Controls agent behavior by using deterministic gates and meta-controllers to validate actions and manage interactions with specialized systems before external execution.
  • Agent Execution Environments - Provides sandboxed runtimes that link agent logic to external APIs and system commands for real-world interaction.
  • Unified Model Interfaces - Provides a standardized factory function to connect to various large language model providers, simplifying how applications request and receive data.
  • Agent Performance Benchmarks - Provides a standardized suite of tasks to compare the effectiveness of different design patterns and quantify accuracy in problem-solving scenarios.

Historique des stars

Graphique de l'historique des stars pour fareedkhan-dev/all-agentic-architecturesGraphique de l'historique des stars pour fareedkhan-dev/all-agentic-architectures

Recherche par IA

Explorez plus de dépôts awesome

Décrivez vos besoins en langage naturel — l'IA classe des milliers de projets open source sélectionnés par pertinence.

Start searching with AI

Collections incluant All Agentic Architectures

Sélections manuelles où All Agentic Architectures apparaît.
  • Raisonnement et planification pour agents LLM
  • une collection de compétences et plugins pour claude code
  • Framework d'orchestration multi-agents

Questions fréquentes

Que fait fareedkhan-dev/all-agentic-architectures ?

Ce projet est un framework complet pour construire, évaluer et connecter des systèmes d'agents autonomes. Il fournit une bibliothèque de modèles architecturaux standardisés pour implémenter des workflows d'agents complexes, y compris l'orchestration multi-agents, le raisonnement itératif et la gestion de la mémoire. En offrant une interface unifiée pour les fournisseurs de modèles, le framework permet une exécution cohérente des agents à travers différents services…

Quelles sont les fonctionnalités principales de fareedkhan-dev/all-agentic-architectures ?

Les fonctionnalités principales de fareedkhan-dev/all-agentic-architectures sont : Agentic LLM Frameworks, Agentic Reasoning Loops, AI Agent Architectures, Autonomous Agent Patterns, External Tool Integrations, LLM Provider Interfaces, Execution Contracts, Memory Storage and Retrieval Systems.

Quelles sont les alternatives open-source à fareedkhan-dev/all-agentic-architectures ?

Les alternatives open-source à fareedkhan-dev/all-agentic-architectures incluent : kyegomez/swarms — Swarms is a multi-agent orchestration framework and autonomous agent toolkit designed to coordinate large language… agiresearch/aios — AIOS is an LLM agent operating system and orchestration kernel designed to manage memory, resource scheduling, and… agentscope-ai/agentscope — Agentscope is a comprehensive toolkit for developing and orchestrating autonomous multi-agent systems. It provides a… modelengine-group/nexent — Nexent is an enterprise AI control plane and LLM agent orchestration platform. It provides a zero-code environment for… panaversity/learn-agentic-ai — This project is an educational curriculum and architectural framework for building autonomous AI agents and… i-am-bee/beeai-framework — The BeeAI Framework is an LLM agent framework and multi-agent orchestration engine used to build autonomous agents…

Alternatives open source à All Agentic Architectures

Projets open source similaires, classés selon le nombre de fonctionnalités partagées avec All Agentic Architectures.
  • kyegomez/swarmsAvatar de kyegomez

    kyegomez/swarms

    6,888Voir sur GitHub↗

    Swarms is a multi-agent orchestration framework and autonomous agent toolkit designed to coordinate large language model agents. It serves as a workflow engine for managing agent relationships, providing the infrastructure to build autonomous agents with integrated memory, tool-calling capabilities, and reasoning loops. The framework is distinguished by its multi-agent consensus systems, which utilize voting, adversarial debates, and judge agents to synthesize high-quality responses. It supports a variety of collaboration patterns, including director-worker hierarchies, expert synthesis, and

    Python
    Voir sur GitHub↗6,888
  • agiresearch/aiosAvatar de agiresearch

    agiresearch/AIOS

    5,168Voir sur GitHub↗

    AIOS is an LLM agent operating system and orchestration kernel designed to manage memory, resource scheduling, and tool execution for multiple autonomous AI agents. It serves as a comprehensive framework for developing and deploying agents, featuring a dedicated resource manager that coordinates model backends, GPU memory, and isolated kernel instances. The system distinguishes itself through a semantic memory engine that uses vector search and autonomous clustering for long-term knowledge management, and a semantic file system that allows users to control computer files and system operations

    Python
    Voir sur GitHub↗5,168
  • agentscope-ai/agentscopeAvatar de agentscope-ai

    agentscope-ai/agentscope

    26,895Voir sur GitHub↗

    Agentscope is a comprehensive toolkit for developing and orchestrating autonomous multi-agent systems. It provides a unified framework for building agents that can reason, execute tools, and manage memory, enabling the creation of complex, collaborative workflows where multiple specialized agents interact to solve multi-step objectives. The platform distinguishes itself through a robust orchestration engine that supports both sequential and concurrent agent pipelines. It utilizes a centralized event bus for real-time telemetry, allowing developers to track agent reasoning, tool usage, and sys

    Pythonagentchatbotlarge-language-models
    Voir sur GitHub↗26,895
  • modelengine-group/nexentAvatar de ModelEngine-Group

    ModelEngine-Group/nexent

    5,265Voir sur GitHub↗

    Nexent is an enterprise AI control plane and LLM agent orchestration platform. It provides a zero-code environment for designing, deploying, and managing production AI agents through a multi-agent collaboration framework that coordinates specialized autonomous agents using standardized messaging protocols. The platform integrates the Model Context Protocol to connect agents with external tools, plugins, and services via a universal communication interface. It further distinguishes itself with a dedicated RAG knowledge base manager that imports unstructured documents and utilizes hybrid search

    Pythonagentagentic-aiagentic-framework
    Voir sur GitHub↗5,265
  • Voir les 30 alternatives à All Agentic Architectures→