Skip to content

Best Self-Hosted AI Agent Platforms Tools

Self-hosted AI agent platforms let you build and run LLM-powered agents and workflows on your own server, so prompts, documents and provider keys stay under your control. The seven picks here sit at different levels. Flowise, Langflow and Sim are visual builders: you wire models, vector stores, tools and integrations together as nodes on a canvas and run the result; Flowise is Node.js with a SQLite default, Langflow is Python and can also serve each flow as an MCP server, and Sim is a Next.js app with a long integration list and an Apache-2.0 core whose ee/ directory carries commercially licensed enterprise modules. Flowise's published code also carries commercially licensed enterprise modules. Dify is the larger application platform, with a visual workflow editor, a built-in RAG pipeline over your own documents and many model connectors, under a modified Apache 2.0 licence that restricts multi-tenant hosting and removing its logo. MaxKB is built around the knowledge base: you upload documents or crawl pages, it vectorizes them and an agent answers from them, and it runs as one GPL-3.0 container. Agent Zero is different again: a general-purpose agent, MIT-licensed, that works in a Linux environment of its own rather than a flow you draw. AgentArea is the infrastructure end: agents run durably on Temporal with network isolation, tool-approval workflows and audit trails, rather than as a library you import. Decide whether you want a canvas, a full app platform, a knowledge base, a general agent or a governed runtime, check the licence terms if you will resell or white-label it, and size the server to match: Dify, AgentArea and Langflow want 2 to 4 GB, MaxKB's docs state 8 GB, Sim's docs list 12 GiB as an evaluation example, Agent Zero's guide gives 2 GB as a minimum, and Flowise needs less.

We earn commissions when you shop through the links below. Full disclosure →

AI Agent Platforms apps

AI Agent Platforms

Agent Zero

A general-purpose AI agent that works inside its own Linux environment in a Docker container, driven from a web interface. The agent can run terminal commands and code, browse the web with a built-in browser you can annotate, use a Linux desktop and LibreOffice documents shown in a canvas beside the chat, delegate work to sub-agents, and keep projects, memory and skills in a persistent user folder. Providers, tools and integrations are added as plugins from a Plugin Hub. It needs a provider API key, a supported AI account or a local model server before it can answer.
PythonFlaskDocker
easyRead guide →
AI Agent Platforms

AgentArea

A self-hosted runtime for governed agentic networks: instead of an agent library you import into your own app, AgentArea runs agents durably as infrastructure, with VPC-style network isolation between them, tool-approval workflows, ReBAC authorization, and full audit trails. Built on Temporal for durable execution, provider-agnostic via LiteLLM, and extensible through hosted MCP servers.
PythonNext.jsGoTemporal
hardRead guide →
AI Agent Platforms

Dify

A platform for building LLM applications — chatbots, agents and multi-step workflows — in a visual editor, with a built-in RAG pipeline over your own documents and connectors for many model providers. Its modified Apache 2.0 license forbids running it as a multi-tenant service or removing the Dify logo from the web frontend without a commercial license.
PythonFlaskNext.jsTypeScript
easyRead guide →
AI Agent Platforms

Flowise

A visual builder for LLM applications and agents: drag nodes onto a canvas to build chatflows, multi-agent flows and assistants, connect them to chat models, vector stores, tools and document loaders, store credentials in the app, and call or embed the result through an API or chat widget. A Node.js server with a React UI that keeps its data in SQLite by default, or in PostgreSQL, MySQL or MariaDB.
TypeScriptNode.jsReact
easyRead guide →
AI Agent Platforms

Langflow

A visual builder for AI agents and workflows: connect models, vector stores and tools as components on a canvas, then serve each flow through an API or as an MCP server. Runs on SQLite by default, or PostgreSQL for production.
PythonFastAPIReactTypeScript
easyRead guide →
AI Agent Platforms

MaxKB

A platform for building question-answering agents over your own documents. You upload files or crawl web pages, MaxKB splits the text and stores vectors for retrieval, and agents answer from that knowledge base using the model you connect. It also has a workflow editor, a function library, MCP tool use and multimodal input and output, and a chat can be embedded into another system without code. It supports private models (DeepSeek, Llama, Qwen) and public ones (OpenAI, Claude, Gemini, MiniMax). The Docker image runs the web app, a task worker and PostgreSQL with pgvector in one container.
PythonDjangoVue.jsPostgreSQL
easyRead guide →
AI Agent Platforms

Sim

A workspace for building, deploying and monitoring AI agents and workflows. You assemble agents on a visual flow editor, connect them to a long list of integrations (Slack, Stripe, Notion, Salesforce, Supabase and many more) and to MCP servers, and run them from the app. It works with hosted models and with local ones through Ollama or vLLM. The self-hosted stack is Next.js on Bun with PostgreSQL and pgvector, Redis for pub/sub and caches, a Socket.io realtime service for the editor and a one-shot migrations job. Features such as single sign-on, access control groups and SCIM sit in an ee/ directory under a separate Enterprise license; the published image contains that code, which stays off unless the enterprise feature flags are set.
TypeScriptNext.jsBunPostgreSQLRedis
easyRead guide →

Head-to-heads in this category

Where to host itaffiliate disclosure
Kamaterafree trial
From $4/mo · 1 vCPU / 1 GB / 20 GB · US + EU + Asia
Start free on Kamatera → (opens in new tab)
DigitalOceanalso works on
From $6/mo · 1 vCPU / 1 GB / 25 GB · US + EU + Asia
Deploy on DigitalOcean → (opens in new tab)

Paid link — we earn a commission if you shop through it.

Replace a paid tool

Common questions

How much RAM do self-hosted apps in AI Agent Platforms need?

Across the 7 apps in AI Agent Platforms, minimums run from 1024 MB (Flowise) to 8192 MB (MaxKB).

Which app in AI Agent Platforms uses the least RAM?

Of the 6 apps in AI Agent Platforms we measured, Flowise used the least: 697 MB at idle (first boot, measured Nov 2026).

Do self-hosted apps in AI Agent Platforms support Docker?

All 7 apps in AI Agent Platforms list Docker support.

Analytics cookies only (Google Analytics, PostHog). No ad networks, no cross-site tracking. See our privacy policy.