OpenHands (fka OpenDevin)

OpenHands agents can do anything a human developer can: modify code, run commands, browse the web, call APIs, and yes—even copy code snippets from StackOverflow.

Agent Framework

Visit Website

Founded

2024

Funding

OSS

OpenHands (formerly OpenDevin) — Open-Source Developer Agent by All Hands AI

OpenHands is an open-source, model-agnostic developer agent that automates real engineering work—code review chores, small bug fixes, refactors, tests, and lightweight feature development. Built and maintained by All Hands AI, it runs agents with real tools (shell, code editor, browser, APIs) to plan tasks, edit files, run commands, browse docs, and open pull requests.

Official site: [All Hands AI](https://www.all-hands.dev/)

Docs and product: [OpenHands Documentation](https://docs.all-hands.dev/)

GitHub repo: [OpenHands on GitHub](https://github.com/All-Hands-AI/OpenHands)

Cloud app: [OpenHands Cloud](https://app.all-hands.dev/)

Research: [arXiv: An Open Platform for AI Software Developers as Generalist Agents](https://arxiv.org/abs/2407.16741)

Funding: $5M seed led by Menlo Ventures (2024) — [coverage](https://techcrunch.com/2024/09/05/all-hands-ai-raises-5m-to-build-open-source-agents-for-developers/)

Why OpenHands

Model-agnostic and OSS: A community-driven alternative to closed agents like Devin.

Real-tool autonomy: Uses a shell, code editor, and browser for practical, repo-level work.

Flexible deployment: Start fast in the cloud, then move on‑prem or VPC for control and compliance.

Transparent operation: A step-by-step agent UI builds trust and simplifies debugging.

What It Does

Addresses PR feedback across multiple files and opens PRs automatically.

Refactors legacy modules and updates dependencies.

Fixes failing tests and expands coverage.

Scaffolds small features and services.

Triages issues and automates repo housekeeping (linting, formatting, minor config fixes).

How It Works

Connect your repo (GitHub, GitLab, Bitbucket), define a task, and set acceptance criteria.

The agent plans, edits, executes shell commands, browses docs, and validates outcomes.

Runs execute in secure, disposable environments backed by [Daytona](https://openhands.daytona.io/).

Integrates with models from OpenAI, Anthropic, or local providers; optional web search for lookups.

Learn more in the [OpenHands docs](https://docs.all-hands.dev/) and the [platform paper on arXiv](https://arxiv.org/abs/2407.16741).

Hosting and Deployment Options

Self-host via Docker with your preferred models and optional search integrations. See the [Quick Start](https://docs.all-hands.dev/openhands/usage/quick-start).

Use OpenHands Cloud for fastest onboarding and managed infra: [OpenHands Cloud](https://app.all-hands.dev/).

Deployment guidance encourages starting in cloud, then migrating on‑prem when you need tighter data and model control.

Integrations

VCS: GitHub, GitLab, Bitbucket

ChatOps: Slack triggers for tasks initiated from conversations

Infra: [Daytona](https://openhands.daytona.io/) secure, disposable environments

Models: OpenAI, Anthropic, and local (e.g., Ollama/LM Studio via docs)

Web: Optional search tools for in-task lookups

Pricing and Free Trial

OpenHands Cloud offers $20 in free credits for new users. Get started via the [Quick Start guide](https://docs.all-hands.dev/openhands/usage/quick-start).

Who It’s For

Software teams seeking an OSS, model-agnostic coding agent.

Staff/platform engineers offloading repetitive engineering toil.

Teams requiring on‑prem or VPC deployment with their own models and data.

Early adopters comfortable supervising agents and tuning prompts/workflows.

Strengths and Tradeoffs

Pros (from user sentiment):

Handles routine dev toil well (e.g., PR feedback, minor bug fixes) — community discussion on [Reddit](https://www.reddit.com/r/ChatGPTCoding/comments/1icbrua/your_thoughts_on_openhands_formerly_opendevin/).

Transparent, step-by-step UI improves trust and debugging — see the [GitHub repo](https://github.com/All-Hands-AI/OpenHands).

Works with real tools across repos (shell, editor, browser), not just chat — detailed in the [docs](https://docs.all-hands.dev/).

Daytona-backed environments reduce setup friction and isolate runs — [Daytona overview](https://openhands.daytona.io/).

Fast-moving maintainers actively engage and iterate — example [HN comment](https://news.ycombinator.com/item?id=42734681).

Cons (from user sentiment):

Token usage can spike; agents may loop/backtrack on harder tasks — reports on [Reddit](https://www.reddit.com/r/ChatGPTCoding/comments/1icbrua/your_thoughts_on_openhands_formerly_opendevin/).

Vulnerable to prompt injection without proper sandboxing/scope — background analysis by [Simon Willison](https://simonw.substack.com/p/prompt-injections-as-far-as-the-eye).

Benchmark gains may not always translate to complex multi-repo production work — [HN discussion](https://news.ycombinator.com/item?id=42734681).

Best Practices for Safe and Effective Use

Scope permissions tightly; prefer ephemeral branches and PR-based changes.

Start with smaller, bounded tasks and clear acceptance criteria.

Set token budgets and monitor spend, especially in cloud runs.

Keep agents sandboxed; validate outputs before merging.

Choose models based on task complexity and cost profile; test local vs hosted LLMs.

Getting Started

1. Read the [Quick Start guide](https://docs.all-hands.dev/openhands/usage/quick-start).

2. Connect your repo (GitHub, GitLab, or Bitbucket) and Slack if desired.

3. Configure model credentials (OpenAI, Anthropic, or local) and optional web search.

4. Define a task with explicit scope, constraints, and success criteria.

5. Review the agent’s step-by-step plan and outputs; merge PRs after validation.

Key Resources

Website: [All Hands AI](https://www.all-hands.dev/)

Product/docs: [OpenHands Documentation](https://docs.all-hands.dev/)

GitHub: [OpenHands on GitHub](https://github.com/All-Hands-AI/OpenHands)

Cloud: [OpenHands Cloud](https://app.all-hands.dev/)

Research: [arXiv paper](https://arxiv.org/abs/2407.16741)

Funding coverage: [TechCrunch](https://techcrunch.com/2024/09/05/all-hands-ai-raises-5m-to-build-open-source-agents-for-developers/)

Community sentiment: [Reddit thread](https://www.reddit.com/r/ChatGPTCoding/comments/1icbrua/your_thoughts_on_openhands_formerly_opendevin/), [HN discussion](https://news.ycombinator.com/item?id=42734681)

Security context: [Prompt injection overview](https://simonw.substack.com/p/prompt-injections-as-far-as-the-eye)

Related Companies

AgentGPT

AgentGPT is an open‑source, browser‑run tool from Reworkd that lets you name an agent, set a goal, and have it autonomously plan, research, and execute tasks in iterative loops. It’s designed for fast experimentation and demos, with ready‑made templates for research, branding, and trip planning—no engineering setup required.

BabyAGI

BabyAGI is pioneering the future of autonomous AI through an experimental framework designed for self-building agents. Born from the insight that the most effective path to general autonomous agents is radical simplicity, BabyAGI focuses on creating the minimal viable system capable of building and evolving itself. At its core is functionz, an innovative function framework that revolutionizes how autonomous agents manage their capabilities. This database-driven system stores, manages, and executes functions through an intelligent graph-based architecture that tracks imports, dependencies, and authentication—all with automatic loading and comprehensive logging. BabyAGI provides developers with an intuitive dashboard for seamless function management, real-time updates, and detailed log analysis, making autonomous agent development accessible and transparent. By embracing a self-building philosophy, BabyAGI represents a fundamentally new approach to creating AI systems that can adapt, grow, and improve autonomously.

CrewAI

CrewAI is at the forefront of Agentic AI with its open source, multi-agent framework and cloud platform for building, managing and scaling agentic workflows across the entire organization.

Flowise AI

Flowise is an open source drag & drop tool to build your customized LLM flow. We provide a visual interface to let you build backends for LLM apps used for QnA, summarization and analysis on your documents.

LangChain

LangChain provides the agent engineering platform and open source frameworks developers need to ship reliable agents fast.

LlamaIndex

LlamaIndex empowers developers to build agents that extract insights and take action on complex enterprise documents. It combines industry-leading document parsing and extraction with a trusted framework for building intelligent agents that reason over documents, adapt to business logic, and scale to production. LlamaIndex is loved by developers and trusted by enterprises. Its open source framework is downloaded more than 4M+ every month and has processed more than 200 million documents on LlamaCloud.