DeerFlow

What is DeerFlow?

DeerFlow is an open-source SuperAgent framework with two layers: DeerFlow Harness, a Python SDK and runtime for building your own long-running agents, and DeerFlow App, a ready-to-deploy application built on top of it. The runtime supplies skills that load progressively, tool use, memory, subagent orchestration, context management and a Docker-based sandbox where the agent can run commands, manage files and execute tasks lasting minutes to hours. The App adds a browser workspace with threads, artifact previews, file uploads, custom named agents, an extensions screen for MCP servers and skills, and a FastAPI gateway. You can self-host it locally with make dev, with Docker Compose, or with Kubernetes-managed sandboxes, and use Postgres or SQLite for state. Sample tasks on the site include deep research reports, data analysis of the Titanic dataset, comics and video generation. The project is MIT licensed, so there is no license fee, though model APIs and hosting are your own cost.

Use cases

• Produce deep research reports
• Run exploratory data analysis with charts
• Execute code in an isolated sandbox
• Summarize podcasts and videos into reports
• Build a custom agent on the Python SDK
• Self-host a team agent workspace

Who it's for

• AI developers
• Researchers
• Engineering teams
• Platform teams

Why use DeerFlow

• Agents run commands in an isolated sandbox rather than directly on your machine.
• The same project gives you a finished app to deploy and an SDK to build something custom.
• Skills can be added as files, so new abilities need no change to the core runtime.
• Self-hosting keeps data handling and runtime configuration under your own control.

DeerFlow features

• Python SDK harness for custom agent systems
• Web workspace with threads and artifact viewer
• Custom named agents with their own models and skills
• Progressively loaded agent skills
• Docker-based AIO Sandbox with browser, shell and files
• Subagent orchestration and memory
• Extensions UI for MCP servers and skills
• FastAPI gateway with LangGraph-compatible runtime
• Sandbox options: local, Docker, E2B cloud or Kubernetes
• Postgres or SQLite state persistence

DeerFlow integrations

• LangGraph
• LangChain
• Docker
• Kubernetes
• E2B
• PostgreSQL
• SQLite
• OpenAI
• MCP servers

How to get started with DeerFlow

1. Clone the repository and create a config.yaml with your model settings and a .env file with your API keys.
2. Run make dev for local use, or docker compose for a containerized setup.
3. Open the app at localhost:2026 and start a thread with a research, coding or creation goal.
4. Choose or add skills, and enable MCP servers from the extensions screen.
5. For many users, switch to a container sandbox and a Postgres database.

DeerFlow limitations

• Self-hosted: you run Docker, nginx and database infrastructure yourself
• SQLite is not suitable for concurrent multi-user production use; Postgres is advised
• Local sandbox is only for single-user, trusted workflows
• Version 2.0 is a rewrite and shares no code with 1.x

DeerFlow pricing

DeerFlow is open source, so you can use and self-host it for free.

Prices can change and may vary by country or exclude tax. Check the official pricing page.

DeerFlow alternatives

The main DeerFlow alternatives listed on Debuly are Manus, LangChain Deep Agents, CrewAI, AutoGPT, OpenHands and AI Manus. The table compares pricing and platforms; order is A–Z.

1. 

   AI Manus

   Open-source, self-hostable general-purpose AI agent that plans tasks and runs a terminal, browser and files inside a Docker sandbox.

   Automation and agentsDeveloper platformsProductivity
   Open sourceVisit AI Manus (official site)
2. 

   AutoGPT

   Agent platform with a plain-English AutoPilot, a visual workflow builder and self-hosting, plus a $50/month Pro plan with a 7-day trial.

   Automation and agentsDeveloper platformsProductivity
   FreemiumVisit AutoGPT (official site)
3. 

   CrewAI

   Multi-agent framework and platform for building AI agent crews in Python or a visual editor, then deploying and monitoring them.

   Developer platformsAutomation and agentsCoding
   Open sourceVisit CrewAI (official site)
4. 

   LangChain Deep Agents

   Open-source LangChain agent harness: long-term memory, a virtual filesystem, subagents and planning included

   Developer platformsAutomation and agentsCoding
   Open sourceVisit LangChain Deep Agents (official site)
5. 

   Manus

   General AI agent that carries out multi-step tasks such as research, slides, websites and design from a single prompt.

   Automation and agentsChat assistantsProductivity
   FreemiumVisit Manus (official site)
6. 

   OpenHands

   Open-source AI coding agent platform with a free Agent Canvas, a Teams plan, and self-hosted Enterprise control

   CodingAutomation and agentsDeveloper platforms
   FreemiumVisit OpenHands (official site)

DeerFlow FAQ

Sources & verification

Facts on this page come from the official sources below, each with the date we read it. Written by Debuly; the vendor did not review this page.

Image credits

• Logo: from deerflow.tech · fetched 23 Oct 2026 · Vendor trademark, used only to identify the product; removed on request
• Cover: none collected yet, so a generated tile is shown.

Spot an error? Report an issue with this listing. We also remove a logo or image on the owner's request.