Category leader in multi-agent orchestration — 69,352 stars (verified), $18.8M Series A, AMD hardware partnership, 455 contributors, 1M downloads/month PyPI (3.4M all-time). SWE-Bench Verified 72% with Claude 4.5 Extended Thinking (updated 2026-03-19), Multi-SWE-Bench #1 across 8 languages. Gap to #2 is enormous on every axis.
Shipped last month
OpenHands drops to #7 in coding-clis (was #5)
No evidence of widespread individual developer adoption despite 69.7K stars and strong benchmarking (OpenHands Index). Low HN traction and cloud-first latency weigh it down. · 2026-03-25
Where it wins
68,800+ stars (verified), 455 contributors — largest open-source community by 10x
4M+ downloads, accelerating
$18.8M Series A (Madrona-led) + AMD Lemonade collaboration
ICLR paper accepted 2025 — peer-reviewed research backing
MIT license, model-agnostic, self-hostable — only serious option for air-gapped deployment
Planning Agent v1.5.0 (Mar 2026) — Plan/Code mode toggle
Biweekly releases (v1.5.0 on 2026-03-11)
Where to be skeptical
SWE-bench Verified 43.2% (CodeAct) — significant gap below closed-source leaders
No disclosed revenue or paying customer count
HN engagement low (<100 pts) compared to peers — 69K stars don't translate to developer mindshare
Enterprise adoption evidence thin beyond logo slides
Development velocity plateaued relative to Codex CLI (896/mo) and Gemini CLI (709/mo)
Editorial verdict
Category leader. No other tool has the combination of open-source community (69K stars, 455 contributors), real download volume (1M/month), venture backing ($18.8M), hardware partnerships (AMD), and benchmark leadership (SWE-bench Verified 72%). Gap to #2 is enormous.
Source
Videos
Reviews, tutorials, and comparisons from the community.
OpenHands + Devstral = A Fully Local Coding Agent
Build Apps With Prompts! - OpenHands
OpenHands: Open-source AI Software Development Agents
OpenHands: AI Software Engineer - Generate Complex Apps
OpenHands: BEST AI Software Engineer
Related
Coding CLIs / Code Agents
Sandboxed agent execution, security-conscious environments, model flexibility without vendor lock-in
Teams of Agents / Multi-Agent Orchestration
End-to-end autonomous coding platform — self-hostable, model-agnostic, enterprise-validated
Software Factories
Regulated industries needing on-prem/air-gapped deployment with MIT-licensed, model-agnostic agents
Claude Code
98Anthropic's official agentic coding CLI. v2.1.81 (Mar 20) shipped `--bare`, smarter worktree resume, and improved MCP OAuth while the repo crossed 82,204 stars and logged ~14 commits/week across 10+ maintainers. Terminal-native, tool-use-driven, with deep file system + shell access, #1 SWE-bench Pro standardized (45.89%), ~4% of GitHub public commits (SemiAnalysis), $2.5B annualized revenue. 8M+ npm weekly downloads. Opus 4.6 with 1M context.
LangGraph
95#1 Python agent framework by production evidence — 40.2M PyPI downloads/month, Fortune 500 deployments (LinkedIn, Uber, Replit, Elastic, Klarna, Cloudflare, Coinbase), ~400 LangGraph Platform companies, LangSmith rated best-in-class observability. Stable v1.x API, model-agnostic, MCP support.
Pydantic AI
95#3 Python agent framework by downloads — 15.6M PyPI/month. Built by the Pydantic team. Runtime type enforcement is a genuine differentiator no other framework offers. V1 shipped with Temporal integration for durable execution and Logfire observability. Emerging pattern: 'Pydantic AI for agent logic, LangGraph for orchestration' (ZenML).
AutoGen (Microsoft)
95⚠️ MAINTENANCE MODE — Microsoft officially confirmed bug fixes and security patches only, no new features (VentureBeat 2026-02-19). 55.9K stars but only 1.57M PyPI/month — DL/star ratio of 28, the most inflated among active frameworks. Being replaced by Microsoft Agent Framework (AutoGen + Semantic Kernel merge, GA targeted ~Q2 2026). Teams on AutoGen should plan migration.
Public evidence
Cloud coding agents that solve 87% of bug tickets same day. 77.6% SWE-bench Verified. New software-agent-sdk v1.0 architecture.
Only multi-agent platform with a major hardware vendor partnership for local inference. Advances agent performance and enterprise readiness utilizing AMD Lemonade Server.
Engineers at AMD, Apple, Google, Amazon, Netflix, TikTok, NVIDIA, Mastercard, and VMWare have cloned or forked the repository. Series A gives runway for cloud product development.
Early enterprise adopters report reducing code-maintenance backlogs by up to 50% and cutting vulnerability resolution times from days to minutes. Transition from research tool to production platform.
Listed among top 18 open-source AI agent projects by GitHub stars. Positioned as the leading autonomous software engineering agent.
Peer-reviewed conference paper at ICLR 2025. Evaluated across 15 benchmarks including SWE-bench and WebArena. Documents 2.1K+ contributions from 188+ contributors.
One tester reports agent completed half a test-writing task but cost $50 in API credits. Maintainer acknowledges it's 'useful for a handful of one-off tasks' but not yet mission-critical.
Tested on real project (Quarto blog migration). Found it effective but not fully autonomous — 'saved me time and mental energy' but needed 'a bit of human guidance.' Transparent about both strengths and limitations.
Used OpenHands as scaffolding framework to generate 67K trajectories for fine-tuning. Describes OpenHands as 'one of the most widely adopted open-source agent scaffolding frameworks.' Achieved 61.7% SWE-bench Verified.
Head-to-head comparison. OpenHands wins on enterprise features (RBAC, audit logs, CI/CD), web UI, and team collaboration. SWE-agent wins on benchmark credibility and academic foundation.
Stars confirm community traction far beyond any competitor in the autonomous coding category. Consistently ranked #1 open-source Devin alternative across multiple independent roundups.
Raw GitHub source
Constrained peek so you can sanity-check the source material without leaving the site.
<a name="readme-top"></a>
<div align="center"> <img src="https://raw.githubusercontent.com/OpenHands/docs/main/openhands/static/img/logo.png" alt="Logo" width="200"> <h1 align="center" style="border-bottom: none">OpenHands: AI-Driven Development</h1> </div> <div align="center"> <br/> <!-- Keep these links. Translations will automatically update with the README. --><a href="https://www.readme-i18n.com/OpenHands/OpenHands?lang=de">Deutsch</a> | <a href="https://www.readme-i18n.com/OpenHands/OpenHands?lang=es">Español</a> | <a href="https://www.readme-i18n.com/OpenHands/OpenHands?lang=fr">français</a> | <a href="https://www.readme-i18n.com/OpenHands/OpenHands?lang=ja">日本語</a> | <a href="https://www.readme-i18n.com/OpenHands/OpenHands?lang=ko">한국어</a> | <a href="https://www.readme-i18n.com/OpenHands/OpenHands?lang=pt">Português</a> | <a href="https://www.readme-i18n.com/OpenHands/OpenHands?lang=ru">Русский</a> | <a href="https://www.readme-i18n.com/OpenHands/OpenHands?lang=zh">中文</a>
</div> <hr>There are a few ways to work with OpenHands:
The SDK is a composable Python library that contains all of our agentic tech. It's the engine that powers everything else below.
Define agents in code, then run them locally, or scale to 1000s of agents in the cloud.
Check out the docs or view the source
The CLI is the easiest way to start using OpenHands. The experience will be familiar to anyone who has worked with e.g. Claude Code or Codex. You can power it with Claude, GPT, or any other LLM.
Check out the docs or view the source
Use the Local GUI for running agents on your laptop. It comes with a REST API and a single-page React application. The experience will be familiar to anyone who has used Devin or Jules.
Check out the docs or view the source in this repo.
This is a deployment of OpenHands GUI, running on hosted infrastructure.
You can try it for free using the Minimax model by signing in with your GitHub or GitLab account.
OpenHands Cloud comes with source-available features and integrations:
Large enterprises can work with us to self-host OpenHands Cloud in their own VPC, via Kubernetes. OpenHands Enterprise can also work with the CLI and SDK above.
OpenHands Enterprise is source-available--you can see all the source code here in the enterprise/ directory, but you'll need to purchase a license if you want to run it for more than one month.
Enterprise contracts also come with extended support and access to our research team.
Learn more at openhands.dev/enterprise
Check out our Product Roadmap, and feel free to open up an issue if there's something you'd like to see!
You might also be interested in our evaluation infrastructure, our chrome extension, or our Theory-of-Mind module.
All our work is available under the MIT license, except for the enterprise/ directory in this repository (see the enterprise license for details).
The core openhands and agent-server Docker images are fully MIT-licensed as well.
