OpenCode vs OpenAI Codex CLI vs OpenHands

Side by side, as alternatives to Cursor. Rows that differ are highlighted; the strongest value in a row is marked. Unknown means no source yet.

OpenCodeanomalyco/opencode

OpenAI Codex CLIopenai/codex

OpenHandsOpenHands/OpenHands

Cursorpaid product

Cost and coverage

Price

$0 software; you run it

Free plan with limited AI use; Plus $8/month, Premium $30/month, Super $100/month, Ultra $200/month; third-party models billed separately by usage source ↗

Code in an AI-first editor

Yes

Multi-line tab completion and next-edit prediction

Yes

Let an agent plan and edit across many files

Partly

Yes

Run terminal commands and tests from the agent

Partly

Yes

Ask questions about your whole codebase

Yes

Connect tools and data through MCP

Partly

Yes

Run agents in the cloud in the background

Partly

Yes

Jobs covered

2 of 7 jobs

3 of 7 jobs

4 of 7 jobs

All 7

Popularity and activity (from GitHub)

Stars

219,699Top

130,321

91,499

—

Gain this week

Unknown

—

Signal

Building history

—

Commits / 4 wk

986Top

892

345

—

Latest release

v1.19.3 · 2026-10-16

rust-v0.163.0 · 2026-10-15

v1.30.0 · 2026-10-22Top

—

Commits / 12 wk

2,164

3,490

740

—

People committing / 12 mo

420+

377+

54+

—

Issues opened / closed, 30 days

3742 / 4694

7371 / 1498

464 / 367

—

Facts, with sources on each project page

Runs on my machine?Differs

Yes, CLI and desktop app

Yes, a command-line agent that runs on your computer; cloud models run on OpenAI's servers unless you use --oss

Yes; Agent Canvas and the agent server run on your machine by default, or on your own Docker host, VM or Kubernetes cluster

Hosted by the vendor

Code licenceDiffers

MIT

Apache-2.0

MIT

—

Model weights licenceDiffers

No single bundled model licence. OpenCode connects to models/providers; their individual terms apply.

No separate model weights

—

Commercial useDiffers

Yes, MIT code

Allowed under Apache-2.0

Yes, MIT allows commercial use

—

Paid services needed?Differs

Needs a model provider: local models via Ollama, LM Studio or llama.cpp, or a paid API

Not required with local models (--oss with Ollama or LM Studio) or the limited ChatGPT Free plan; paid ChatGPT plans and OpenAI API keys are optional

Not required with local models (LM Studio, Ollama, vLLM, SGLang); cloud model APIs and OpenHands Cloud are optional

Subscription

HardwareDiffers

No hardware requirements documented in the README or docs (checked 2026-10-21)

Not documented in the README or CLI docs

No requirements given for Agent Canvas itself; the recommended local model needs a 24GB GPU or an Apple Silicon Mac with 64GB memory

—

Operating systemsDiffers

Windows, macOS and Linux; on Windows the docs recommend WSL for the CLI, and there is a Windows desktop app

macOS, Linux and Windows (native PowerShell with a Windows sandbox, or WSL2)

macOS, Linux and Windows (PowerShell or WSL2); the CLI and per-conversation Docker sandboxes need WSL on Windows

—

Setup effortDiffers

Easy; one-line install script, npm, Homebrew, Chocolatey or Scoop, then add a model provider

Easy; one-line install script, npm or Homebrew, then sign in with ChatGPT

Medium; install Node.js 24 and uv then the npm package, or run the Docker image, then add a model

—

Stars over time

Switch to Growth % to compare projects of different sizes fairly.

OpenCode
OpenAI Codex CLI
OpenHands

Focus a point, use arrow keys, or tap for its value.

Tracked daily since 19 Oct 2026. Dashed history: estimated from public records of when people starred the repo. Lines connect observations only; gaps contain no measured points.

Growth % starts at each project's first observed point in the shown range. Zero-star starts are unavailable.

Notes for a video or post

A plain summary with today's date and a link to the sources. Check each project page before you publish.

OpenCode vs OpenAI Codex CLI vs OpenHands (as alternatives to Cursor)
Figures from RepoQuarry, 22 Oct 2026. Sources: https://repoquarry.dev/compare?ids=opencode,codex-cli,openhands&product=cursor

Cursor: Free plan with limited AI use; Plus $8/month, Premium $30/month, Super $100/month, Ultra $200/month; third-party models billed separately by usage
OpenCode: free software; does 2 of 7 Cursor jobs
OpenAI Codex CLI: free software; does 3 of 7 Cursor jobs
OpenHands: free software; does 4 of 7 Cursor jobs

- OpenCode: 219,699 GitHub stars, 2,164 commits in 12 weeks, licence MIT
- OpenAI Codex CLI: 130,321 GitHub stars, 3,490 commits in 12 weeks, licence Apache-2.0
- OpenHands: 91,499 GitHub stars, 740 commits in 12 weeks, licence MIT

Where they differ:
- Runs on my machine?: OpenCode Yes, CLI and desktop app; OpenAI Codex CLI Yes, a command-line agent that runs on your computer; cloud models run on OpenAI's servers unless you use --oss; OpenHands Yes; Agent Canvas and the agent server run on your machine by default, or on your own Docker host, VM or Kubernetes cluster
- Code licence: OpenCode MIT; OpenAI Codex CLI Apache-2.0; OpenHands MIT
- Model weights licence: OpenCode No single bundled model licence. OpenCode connects to models/providers; their individual terms apply.; OpenAI Codex CLI No separate model weights; OpenHands No separate model weights
- Commercial use: OpenCode Yes, MIT code; OpenAI Codex CLI Allowed under Apache-2.0; OpenHands Yes, MIT allows commercial use
- Paid services needed?: OpenCode Needs a model provider: local models via Ollama, LM Studio or llama.cpp, or a paid API; OpenAI Codex CLI Not required with local models (--oss with Ollama or LM Studio) or the limited ChatGPT Free plan; paid ChatGPT plans and OpenAI API keys are optional; OpenHands Not required with local models (LM Studio, Ollama, vLLM, SGLang); cloud model APIs and OpenHands Cloud are optional
- Hardware: OpenCode No hardware requirements documented in the README or docs (checked 2026-10-21); OpenAI Codex CLI Not documented in the README or CLI docs; OpenHands No requirements given for Agent Canvas itself; the recommended local model needs a 24GB GPU or an Apple Silicon Mac with 64GB memory
- Operating systems: OpenCode Windows, macOS and Linux; on Windows the docs recommend WSL for the CLI, and there is a Windows desktop app; OpenAI Codex CLI macOS, Linux and Windows (native PowerShell with a Windows sandbox, or WSL2); OpenHands macOS, Linux and Windows (PowerShell or WSL2); the CLI and per-conversation Docker sandboxes need WSL on Windows
- Setup effort: OpenCode Easy; one-line install script, npm, Homebrew, Chocolatey or Scoop, then add a model provider; OpenAI Codex CLI Easy; one-line install script, npm or Homebrew, then sign in with ChatGPT; OpenHands Medium; install Node.js 24 and uv then the npm package, or run the Docker image, then add a model

Stars show popularity, not quality. Check each project page before publishing.

Want share images and a ready video description? Open a project's creator kit.

Numbers: GitHub, fetched nightly. Facts: read from each project's own documentation, with the source on its page. How we check. anomalyco/opencode ↗openai/codex ↗OpenHands/OpenHands ↗

OpenCode vs OpenAI Codex CLI vs OpenHands

Stars over time

Notes for a video or post

Cookie preferences

Report a problem