Claude Code alternatives in 2026: 8 coding agents and apps to use instead

Disclosure: Enjoy is our product, and it runs Claude Code. We checked every option below against its maker’s own pages on October 28, 2026, and we say where another option is the better pick, Claude Code included.

Claude Code alternatives compared

Claude Code is Anthropic’s coding agent: you describe a change, and it reads your project, edits files and runs commands until it’s done. It runs in a terminal, in Claude’s desktop app and in your editor, on a paid Claude plan or a Console account. The options below do the same job with other labs’ models, on other plans, or with code you can read. One of them, Enjoy, runs Claude Code itself beside the others.

Option	Best for	Price, checked October 28, 2026	Runs on	Models
Claude Code	Claude, from the company that makes it	Claude Pro $20 a month, or $17 billed yearly; Max 5x $100; Max 20x $200	Mac, Windows, Linux	Anthropic’s Claude models
Codex	OpenAI’s models, with a free start	Free and Go ($8) in the desktop app; Plus $20; Pro from $100	Mac, Windows, Linux	GPT-6 Luna on Free and Go; GPT-6.1 Sol from Plus
OpenCode	Free and open source, with any provider	Free; other providers at their own prices	Mac, Windows, Linux	75+ providers, free models and local ones
GitHub Copilot CLI	Several labs’ models in one plan	Free; Pro $10 a month; Pro+ $39; Max $100	Mac, Windows, Linux	Claude, GPT and Gemini; Claude Opus 5.5 from Pro+
Antigravity CLI	Gemini for free, and Claude on Google’s plan	Free; Google AI Pro $19.99 a month; Ultra from $99.99	Mac, Windows, Linux	Gemini; Claude Sonnet and Opus 5.5 on paid Pro and Ultra
Cursor CLI	Grok in a terminal agent, from $8	Free; Plus $8 a month, Grok and Composer only; Premium $30, other labs’ models at API prices	Mac, Windows, Linux	Grok and Composer; Claude, GPT and Gemini billed separately
Kimi Code	A model Moonshot puts close to Opus 5.5	With Kimi Plus, $180 a year, up to Ultra, $1,908 a year	Mac, Windows, Linux	Kimi K3.1 Pro
Cline	An open-source agent in your editor, with your own keys	Free; ClinePass $9.99 a month	VS Code, JetBrains IDEs, Zed and other editors, and a CLI	Any provider’s key
Enjoy	Keeping Claude Code, with other labs’ agents beside it	A monthly subscription, free for 7 days, with your phone and other computers (pricing)	Mac, Windows, Linux	Claude Code, Codex, Grok Build, Antigravity, Muse Code, OpenCode

Prices are per month as each maker lists them in the US, before tax, unless the maker sells by the year. Claude Code, Codex, Copilot, Antigravity, Cursor and Kimi Code come with their makers’ own plans; OpenCode, Cline and Enjoy run on the plans, keys and free models you choose.

Why people look for a Claude Code alternative

Four things on Anthropic’s own pages send people looking:

• There’s no free way in. Its setup guide says “The free claude.ai plan does not include Claude Code access,” so it starts with Pro, at $20 a month. Anthropic’s advice if you “consistently hit limits” is Max, at $100 or $200.
• The limits are shared and unpublished. “Both Pro and Max plans have a five-hour session limit and a weekly limit,” says Anthropic’s help center, and “These limits are shared across Claude and Claude Code, meaning all activity in both tools counts against the same limits.” Its Pro plan page says “The number of messages you can send will vary,” and “we may limit your usage in other ways, such as weekly and monthly caps or model and feature usage, at our discretion.” Its IPO filing adds: “We have also changed, and may in the future change, usage limits, rate limits, or model and product capabilities,” as our memo on the filing explains.
• It’s closed, and it’s Claude. Its license reads “© Anthropic PBC. All rights reserved.” It’s built for Anthropic’s models: the other providers its setup guide names, such as Amazon Bedrock, serve Claude too.
• Windows gets less. On native Windows, its setup guide lists sandboxing as “Not supported”; that takes WSL 2.

Codex

Best for: OpenAI’s models, with a free start and an agent whose code you can read.

OpenAI’s Codex runs in the ChatGPT desktop app, in a terminal and in your editor. Its CLI, a “Lightweight coding agent that runs in your terminal,” is open source under Apache 2.0, and the desktop app is for “macOS, Windows, or Linux.” On Windows, “It runs natively on Windows using PowerShell and the Windows sandbox,” per its Windows guide.

Price: Codex comes with every ChatGPT plan, from Free. Codex’s pricing gives Free, and Go at $8 a month, “GPT-6 Luna at Standard speed in the desktop app, subject to rollout.” Plus, at $20, adds “Codex on the web, in the CLI, in the IDE extension, and on iOS” and GPT-6.1 Sol, and Pro is $100, $200 or $500 a month.

Watch for: OpenAI estimates 15 to 160 local messages every five hours with GPT-6.1 Sol on Plus, but “These estimates are not fixed message limits,” and “Weekly limits may also apply.” “Pro plans currently have no five-hour limit.” It runs only OpenAI’s models, and the free plans get neither Sol nor the CLI. Our Claude Code vs Codex guide compares the two in detail.

OpenCode

Best for: free, open-source coding with the provider you choose, local models included.

OpenCode is “The open source AI coding agent,” under the MIT license on GitHub, where it has more than 220,000 stars. It’s “Available as a terminal interface, desktop app, and IDE extension,” with installs for macOS, Linux and Windows, and the desktop app is in beta. Its homepage lists “75+ LLM providers through Models.dev, including local models,” and you can “Log in with GitHub to use your Copilot account” or “Log in with OpenAI to use your ChatGPT Plus or Pro account.”

Price: “OpenCode is 100% free to use,” and “It also comes with a set of free models.” Other providers bill you at their own prices.

Watch for: your Claude plan doesn’t come with you. OpenCode’s providers page says “There are plugins that allow you to use your Claude Pro/Max models with OpenCode. Anthropic explicitly prohibits this,” so Claude in OpenCode means an Anthropic API key, billed per token. Its free models are free “for a limited time,” and for some of them, “The team is using this time to collect feedback and improve the model.”

GitHub Copilot CLI

Best for: Claude, GPT and Gemini in one plan from $10 a month, in your terminal.

“The command-line interface (CLI) for GitHub Copilot allows you to use Copilot directly from your terminal,” says GitHub, on Linux, macOS, and “Windows from within Powershell and Windows Subsystem for Linux (WSL).” Its plans page says “All plans include Copilot CLI and the Copilot app.” You change models with /model, and you can bring your own: “Use a model from an external provider of your choice in Copilot by supplying your own API key,” from Anthropic, OpenAI, or a local runtime such as Ollama.

Price: Copilot Free costs nothing. Pro is “$10 USD per month (free for some users),” with 1,500 GitHub AI Credits a month and Claude Sonnet 5.5 among its models; Pro+ is $39, with 7,000 credits and Claude Opus 5.5; and Max is $100, with 20,000.

Watch for: “On Copilot Free and Copilot Student plans, access to models is available through auto model selection only,” so you can’t pick Claude there. And “AI credits are consumed based on the number of tokens processed,” so a long session spends more of your month than a short one.

Antigravity CLI

Best for: Gemini for free, and Claude on Google’s plan instead of Anthropic’s.

Google’s Gemini CLI is still open source, but it now says “Free individual access and Google AI Pro / Ultra (Google One) moved to Antigravity CLI on June 18, 2026.” Antigravity CLI, run as agy, installs on macOS, Linux and Windows, and Google’s plans page gives every plan “Access to all product features, such as scheduled tasks and the CLI.”

Price: free. Google AI Pro is $19.99 a month, and Ultra starts at $99.99. Its models page gives a free account Gemini 3.8 Flash, while a paid Pro plan adds Gemini 4 Argon and Claude Sonnet 5.5 and Opus 5.5, “Available on Google AI Pro for non-trial subscriptions only.”

Watch for: without Pro or Ultra, you get “A meaningful quota, refreshed weekly.” Claude Sonnet 4.6 and Opus 4.6, the Claude models on free accounts, “Will be removed on November 2, 2026,” as our memo on the change explains. And there’s no support for “Bring-your-own-key (BYOK) or bring-your-own-endpoint for additional rate limits,” so more usage comes from Google’s plans.

Cursor CLI

Best for: Grok and Cursor’s own models in a terminal agent, from $8 a month.

“Cursor is sold as part of The X Subscription, one subscription for Grok, Cursor, and X,” says its models page. Besides its editor, it has a CLI, run as agent, that installs on “macOS, Linux and Windows (WSL)” and “Windows (native),” and offers “Frontier models from Anthropic, OpenAI, Gemini, Cursor, and more.”

Price: its pricing has Free, with “Limited agent requests”; Plus at $8 a month, where “Only Grok models are available on this plan”; Premium at $30; Super at $100; and Ultra at $200.

Watch for: “Plus and Free don’t include third-party models.” On Premium and above, Claude, GPT and Gemini are paid “at the model’s API list price from the first request,” and Claude Opus 5.5 costs $4 per million input tokens and $20 per million output, as on Anthropic’s own API. Our Cursor alternatives guide covers its plans.

Kimi Code

Best for: a model Moonshot puts within a point of Opus 5.5, in an agent of its own or inside Claude Code.

“Kimi Code is an intelligent programming service for developers included in Kimi membership benefits,” say its docs, “Powered by our latest flagship model, K3.1 Pro.” Its CLI installs on macOS, Linux and Windows and is open source under MIT on GitHub, and there’s a desktop app and a VS Code extension. Moonshot says “K3.1 Pro scores 65.7%, ahead of GPT-6 Astra (58.2%) and narrowly behind Claude Opus 5.5 (66.4%)” on Terminal-Bench 4.0. If you’d rather keep Claude Code’s tools, Kimi’s guide points Claude Code at Kimi instead, as our usage limit guide shows.

Price: Kimi Code comes with Kimi’s memberships from Plus, at $180 a year, shown as $15 a month, up to Ultra, at $1,908 a year. Its membership guide says “The Go tier has no coding quota.”

Watch for: Kimi’s pricing page warns: “K3.1 Pro requests may queue or fail at launch.” Its usage runs on a “rolling 5-hour rate window” and a monthly total, and when that’s spent, “Kimi Code quota is frozen until the monthly quota resets or you upgrade.” The benchmark is Moonshot’s own.

Cline

Best for: an open-source agent inside VS Code, JetBrains IDEs and other editors, with your own keys.

Cline is an “Autonomous coding agent as an SDK, IDE extension, or CLI assistant,” under Apache 2.0 on GitHub. Its docs list VS Code, Cursor, Windsurf, JetBrains IDEs, Antigravity and Zed, and the CLI installs with npm i -g cline. “Cline is not locked to a single AI provider.”

Price: “The Cline open source extension is completely free for individual developers,” says its pricing. “You can bring your own API keys or use the Cline provider,” and ClinePass is a “Flat $9.99/month subscription that offers 2-5x the usage on popular open coding models compared to standard API rate.”

Watch for: you pay for the models separately: by the token with your own keys or Cline’s credits, or with ClinePass for open models. Its pricing names Anthropic among the providers whose keys you can bring, and an API key bills by the token, apart from any Claude plan you pay for.

Enjoy

Best for: keeping Claude Code, with Codex, Grok Build and other labs’ agents ready when it runs out.

Enjoy is a desktop app for Mac, Windows and Linux that runs Claude Code itself, signed in to your Claude plan, along with Codex, Grok Build, Antigravity, Muse Code, and local models through OpenCode. What it adds:

• Hand the conversation on. Switch agents at any point: “The new agent gets the conversation so far, including replies from other agents, before it works on your message.”
• Keep going through limits. When Claude Code runs out, Enjoy switches to another of your Claude accounts with usage left, or waits for the reset and carries on. Select Stop waiting to hand it to Codex or another agent instead.
• Your history, brought in. Enjoy imports a project’s past Claude Code conversations, so you pick up where you left off.
• Every project at once. Each conversation runs its own agent, and JOY keeps an eye on all your projects: ask it what needs you, or start work in any of them.

Price: Enjoy is one monthly subscription, and the first seven days are free, with a card to start. It includes your phone and any browser and your other computers, and each teammate you invite needs a subscription of their own. See pricing.

Watch for: Enjoy doesn’t change Anthropic’s limits or add Claude to another lab’s plan. Claude Code still needs a Claude plan, and Enjoy runs the Codex CLI, which OpenAI includes from Plus. Waiting for a reset needs “Your computer must be on, with Enjoy open, when the usage resets.” Enjoy isn’t open source, it’s a monthly subscription on top of your AI plans once the free week ends, and scheduled workflows run only while your computer is awake with Enjoy open.

When to stay with Claude Code

If Claude does your work and you rarely run out, none of this is a reason to move. Moonshot’s own Terminal-Bench 4.0 figures put Claude Opus 5.5 ahead of K3.1 Pro and GPT-6 Astra. And a limit needn’t end the task: Claude Code’s docs say it “waits in the open session and continues the task on its own after the limit resets,” and Anthropic’s help center says you can “Enable usage credits to keep using Claude with your Pro plan after you reach your five-hour session limit or your weekly limit.” Our usage limit guide compares the ways to keep going.

Move when you want a free start, an agent you can read and change, several labs’ models in one plan, or a sandbox on native Windows. You don’t have to choose only one: Claude Pro and ChatGPT Plus together cost $40 a month, and Enjoy hands a conversation from Claude Code to Codex and back.

How we compared

We read Anthropic’s pricing, help center, Claude Code docs, license and S-1, and each other maker’s homepage, pricing, plans, docs and GitHub repository on October 28, 2026, and every one is linked below. Licenses and star counts come from GitHub that day. We quote each maker’s own words for its limits and leave out what it hasn’t documented, and benchmark figures are the maker’s own. Plans and models in this category change often, so we recheck this guide against its sources at least once a month, and the date at the foot of the page says when we last did.