awesome-ai-stack

Agno

Agno

High-performance framework and runtime (AgentOS) for multi-agent systems

AutoGen

Microsoft

Microsoft's multi-agent conversation framework (maintenance-only; new work moved to Microsoft Agent Framework)

CrewAI

CrewAI

Role-based multi-agent orchestration — assemble a crew of agents with goals and tasks.

DSPy

Stanford NLP

Program — don't prompt — language models; declare modules and let DSPy compile and optimize the prompts.

Structured, validated outputs from any LLM — define a Pydantic model and get a typed object back.

LangChain

LangChain

The original LLM-app framework — standard model/tool interfaces, hundreds of integrations, and the create_agent prebuilt that compiles to a LangGraph agent

LangGraph

LangChain

Graph-based orchestration for stateful, multi-step agents with branching, loops, and human-in-the-loop

LlamaIndex

LlamaIndex

Data framework for building RAG pipelines and document agents over your own data

OpenAI Agents SDK

OpenAI

OpenAI's lightweight SDK for agents with tools, handoffs, and guardrails

Pydantic AI

Pydantic

Type-safe agent framework from the Pydantic team — structured, validated results with any model

smolagents

Hugging Face

Hugging Face's minimal library for agents that act by writing and running code

Aider

Aider

AI pair programming in your terminal — edits your code and auto-commits each change to git.

Cline

Cline

Autonomous coding agent inside VS Code — plans, edits files, runs commands, and supports MCP tools.

Continue

Continue

Open-source AI code assistant for VS Code and JetBrains — autocomplete, chat, and agents

Goose

Block

Block's open-source, extensible coding agent for the terminal and desktop

mini-SWE-agent

Princeton NLP

The minimal successor to SWE-agent — a ~100-line core that autonomously resolves GitHub issues, as performant but far simpler

opencode

SST

SST's open-source, model-agnostic terminal coding agent with a TUI

OpenHands

All Hands AI

Open-source platform for autonomous AI software developers that edit code, run commands, and browse the web

Roomote

Roo Code

Formerly Roo Code

Cloud autonomous coding agent that runs tasks end to end across Slack, GitHub, and Linear

SWE-agent

Princeton NLP

Princeton's agent that autonomously resolves real GitHub issues via an agent–computer interface. Now maintenance-only — superseded by mini-SWE-agent

Anthropic Claude

Anthropic

Frontier LLM API with strong tool-use and long-context reasoning, ideal as an agent's core model

Google Gemini

Google

Google's Gemini models via the AI Studio API — long context, multimodal, with a free tier

Groq

Groq

Ultra-fast inference API for open models on custom LPU hardware

Call 100+ LLM APIs through one OpenAI-style interface, plus a proxy with routing, fallbacks, and cost tracking.

LM Studio

LM Studio

Desktop app to discover, run, and serve local LLMs with an OpenAI-compatible server

Mistral AI

Mistral AI

European model lab with strong open-weight models and a hosted API (La Plateforme)

Ollama

Ollama

Run open LLMs locally with one command — an OpenAI-compatible API for offline, private agents.

OpenAI

OpenAI

OpenAI's GPT models via the API — function calling, structured outputs, and the de-facto API shape

One OpenAI-compatible API to 300+ models across providers, with routing and fallbacks

vLLM

PyTorch Foundation

High-throughput serving engine for running open LLMs with an OpenAI-compatible API

Chroma

Chroma

Open-source embedding database for RAG — store vectors and query by similarity in a few lines.

LanceDB

LanceDB

Embedded vector database built on the Lance columnar format — serverless and local-first

Memory layer for agents — extracts and recalls facts across sessions, not raw chat logs.

Milvus

Zilliz

Open-source vector database built for billion-scale similarity search

Postgres extension for vector similarity search — add agent memory and RAG without a separate database.

Qdrant

Qdrant

High-performance open-source vector database for similarity search and agent memory

Sentence Transformers

Hugging Face

Run embedding models locally (e.g. bge-m3) to turn text into vectors for RAG and memory — no API.

Weaviate

Weaviate

Open-source vector database with hybrid search, filtering, and built-in vectorizers

Zep

Zep

Long-term memory layer for agents, building a temporal knowledge graph from conversations

Arize Phoenix

Arize AI

Open-source LLM and agent observability — OpenTelemetry tracing and evals you can self-host

DeepEval

Confident AI

Open-source "Pytest for LLMs" — unit-test your model and RAG output with metrics.

Helicone

Helicone

Open-source LLM observability via a one-line proxy — logging, caching, and cost tracking

Langfuse

Langfuse

Open-source tracing, evaluation, and analytics for LLM and agent applications

LangSmith

LangChain

LangChain's hosted platform for tracing, evaluating, and monitoring LLM and agent apps (works with or without LangChain)

Opik

Comet

Comet's open-source platform for tracing, evaluating, and monitoring LLM apps

Ragas

Exploding Gradients

Open-source evaluation framework for RAG pipelines and LLM applications

Browser Use

Browser Use

Let an LLM agent drive a real browser — navigate, click, type, and extract from any website.

Composio

Composio

Connects agents to 1000+ external apps with managed auth and ready-made tool definitions.

Secure, elastic infrastructure for running AI-generated code in fast, fully isolated sandboxes

The container runtime most self-hosted agent sandboxes are built on — isolate and run model-generated code in a throwaway container

E2B

E2B

Secure, isolated cloud sandboxes that let AI agents run model-generated code safely

Exa

Exa

Neural web search API for agents — semantic search plus full page contents in one call

Firecrawl

Firecrawl

Crawl or scrape any URL and get back clean, LLM-ready markdown or structured data.

Serverless cloud runtime whose Sandboxes run untrusted, model-generated code in isolated containers on demand

Stagehand

Browserbase

AI browser-automation framework — drive Playwright with act, extract, and observe in natural language.

Web search API built for LLMs and agents — returns clean, ranked, RAG-ready results from one call.

Guardrails AI

Guardrails AI

Validate and structure LLM input/output with composable validators from the Guardrails Hub.

Programmable rails for LLM apps — keep conversations safe, on-topic, and on-policy with Colang flows.

LiveKit Agents

LiveKit

Realtime framework for voice, video, and physical AI agents on the LiveKit WebRTC stack

Pipecat

Daily

Open-source Python framework for real-time voice and multimodal conversational agents