Skip to content

Smithech/awesome-ai-agent

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

24 Commits
 
 
 
 
 
 
 
 
 
 
 
 

Awesome AI Agents Awesome

Build and deploy autonomous and multi-agent systems powered by large language models (LLMs).

Contents

Cloud Platforms for Agents

  • Amazon Bedrock - The AWS platform for building generative AI applications and agents.
  • Vertex AI Agent Builder - A suite of Google Cloud products designed to build, scale, and manage AI agents in production environments.

Context Processing

Embedding models

Transformers

Foundation Models Providers

  • Anthropic Claude - Foundational models such as Haiku, Opus, and Sonnet.
  • Google DeepMind - The Gemini family and Gemma open models, spanning multimodal and lightweight use cases.
  • Meta LLaMA - A family of open-weight language models designed for developers and researchers, supporting fine-tuning, adaptation, and deployment across a broad ecosystem.
  • Open AI - Frontier and specialized models for text, image, speech-to-speech, text-to-speech and transcription tasks.

Inference Providers

  • Cerebras - High-performance AI inference infrastructure focused on large-scale workloads and low-latency execution.
  • Cohere - Enterprise-oriented language models and inference APIs for NLP and retrieval-based applications.
  • Fal - Platform for running and fine-tuning generative media models (image, video, audio) using serverless and on-demand GPU infrastructure.
  • Hyperbolic - Open-access cloud platform for running and serving AI models.
  • Featherless - Infrastructure for deploying and serving open-weight models with minimal setup.
  • Fireworks - Inference platform for open-source models with optimization for performance, scalability, and customization.
  • Groq - Low-latency inference platform powered by custom hardware for deterministic model execution.
  • HF Inference - Serverless inference APIs provided by Hugging Face for deploying and consuming machine learning models.
  • Novita - Unified API platform for accessing and deploying multiple models and running agent-based workflows.
  • Nscale - Infrastructure provider covering compute, storage, and deployment for AI systems across environments.
  • ovhOVH AI Endpoints - Managed APIs for integrating and serving machine learning and generative AI models.
  • Public AI - Open-source and nonprofit initiative providing shared infrastructure for public AI model access and experimentation.
  • Replicate - Platform for running, deploying, and fine-tuning models via API-based workflows.
  • SambaNova - AI inference systems built on specialized hardware and software for large-scale model execution.
  • Scaleway - Cloud platform supporting the deployment and scaling of AI models and applications.
  • Together AI - Platform for training, fine-tuning, and serving open and research-driven AI models.
  • WaveSpeedAI - Infrastructure for accelerating generative media workloads, particularly image and video models.
  • Zai - Platform providing access to conversational AI and agent-based systems.

Interoperability Protocols

  • Agent2Agent (A2A) - An open standard designed to enable seamless communication and collaboration between AI agents.
  • Agent Payments Protocol (AP2) - An open protocol for the emerging Agent Economy. It enables secure, reliable, and interoperable agent commerce for developers, merchants, and the payments industry.
  • Model Context Protocol - (MCP) - An open-source standard for connecting AI applications to external systems.

Local LLM Tools

  • DiffusionBee - Desktop application for running generative models locally, with a focus on image generation.
  • Docker Model Runner - Tooling for managing, running, and deploying AI models within Docker-based environments.
  • Draw Things - Application for running image generation models locally, with support for offline workflows.
  • Jan - Local-first AI assistant designed to run models privately on user devices.
  • JellyBox - Environment for running AI models locally with full offline support.
  • Lemonade - Open-source local AI runtime for deploying and interacting with models on personal hardware.
  • Local AI - Self-hosted AI stack for running language models, agents, and related workloads locally.
  • llama.cpp - Lightweight inference engine in C/C++ for running large language models on local hardware.
  • LM Studio - Desktop interface for discovering, running, and interacting with local language models.
  • MLX LM - Python library for inference and fine-tuning of language models on Apple Silicon using MLX.
  • Ollama - Tool for running and managing language models locally with a simplified CLI and API.
  • SGLang - High-performance framework for serving language and multimodal models.
  • Unsloth - Toolkit for running and fine-tuning models locally, with support for offline environments.
  • vLLM - Inference and serving engine optimized for throughput and memory efficiency in LLM workloads.

Observability

  • LangSmith Platform - Framework-agnostic platform for monitoring, evaluating, and debugging LLM applications and agents.

Orchestration Frameworks

  • Deep Agents - Open-source agent framework for long-running tasks, with support for planning, context management, and multi-agent coordination.
  • Google Agent Development Kit (ADK) - Framework for building AI agents with a model-agnostic and deployment-agnostic design.
  • LangChain - Open-source framework providing abstractions, integrations, and tooling for building LLM-powered applications.
  • LangGraph - Low-level orchestration framework for building and running stateful, long-lived agent workflows.
  • Microsoft Agent Framework - Framework for developing agent-based systems, supporting both simple interactions and multi-agent workflows with graph-based orchestration in .NET and Python.

Sandboxes

  • Amazon Bedrock AgentCore - Managed environment for deploying and running AI agents with support for multiple models and frameworks.
  • Daytona - Infrastructure for executing AI-generated code in isolated and reproducible environments.
  • Modal Sandboxes - Serverless container-based environments for running AI-generated code with support for dynamic configuration and GPU workloads.
  • Runloop - Ephemeral development environments for executing code in isolation, with support for agent-based workflows and evaluation pipelines.

Learning resources

Google

Huggin Face

  • AI Agents Course - This free course will take you on a journey, from beginner to expert, in understanding, using and building AI agents.
  • MCP Course - This free course, built in partnership with Anthropic, will take you on a journey, from beginner to informed, in understanding, using, and building applications with MCP.

LangChain

  • Ambient Agents with LangGraph - Build your own ambient agent to manage your email. You’ll learn the fundamentals of LangGraph as you build an email assistant from scratch, and use LangSmith to evaluate its performance.
  • Building Reliable Agents - Take an agent from first run to production-ready system through iterative cycles of improvement with LangSmith, the agent engineering platform for observing and evaluating agents.
  • Deep Agents - Learn the fundamental characteristics of Deep Agents and how to implement your own Deep Agent for complex, long-running tasks.
  • Deep Research with LangGraph - Build your own deep research agent to handle research tasks. Learn how to use LangGraph to build a multi-agent system, then use LangSmith to evaluate its performance.
  • Introduction to Agent Observability & Evaluations - Learn the essentials of agent observability & evaluations with LangSmith. Continuously improve your agents with LangSmith's tools for observability, evaluation, and prompt engineering.
  • Introduction to LangChain - Python - Learn how to build AI agents with LangChain. Get started quickly using pre-built architectures and model integrations, then debug your agents with LangSmith Observability.
  • Introduction to LangGraph - Python - Learn the basics of LangGraph, the framework helps developers add better precision and control into agentic workflows.
  • Quickstart courses - Collection of quickstart courses about LangChain, LangGraph and LangSmith.

Microsoft

Contributing

Your contributions and suggestions are heartily welcome. Please check the Contributing Guidelines for more details.

About

A curated collection of resources, tools, and frameworks for building, developing, and deploying AI agents. It also provides a list of learning resources about agents, frameworks, MCPs, and everything related to AI agents.

Topics

Resources

License

Code of conduct

Contributing

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors