The argument treats them as three brands of the same product. They are not. Cursor is an editor you work inside. Claude Code is a terminal agent that spends your Claude plan. Codex is the coding agent included with ChatGPT, not a second subscription. Pick the job, then the plan. Data checked 2026-10-19.
A beginner should not start here when the goal is a page or an app they can describe without reading code. No-code weights are 1/5 for Cursor, 0/5 for Claude Code, and 1/5 for Codex. Claude Code assumes a terminal, a repo, and comfort with commands. Use Lovable vs Bolt vs Replit until you can follow a diff.
A landing page, a first web app, or anything you hope to publish without opening files belongs in a browser builder. Cursor's own miss list includes publishing a landing page without looking at code. Codex is a poor pick if you never want to see a file. Claude Code's no-code weight is zero.
Free Claude does not include Claude Code. You need Pro or Max before this tool exists for you. Buying Cursor or ChatGPT Plus does not unlock Claude Code. They are different vendors and different pools.
Cursor Free, the Codex trial, and Claude's paid pool do not publish a fixed prompt count. If a number on a page is how you decide, read Copilot's credit table, and treat anything marked verify as unfinished.
One $20 plan is the experiment. A second $20 plan is for a second job, after the first one is actually how you work. If ChatGPT or Claude is already on your card, use the agent that subscription already includes before you add Cursor.
Codex lists a Go row at "Confirm on chatgpt.com/pricing". Left out of budget matching. The $8 figure is on a Codex pricing doc, and the live ChatGPT pricing page still does not show it. Cursor also lists a regional plan that is left out of dollar matching. Both show on the cards so a rumor cannot become the floor.
Checked 2026-10-19
An editor with agents, built for people who can read a diff.
Cursor is a VS Code-style editor plus a CLI and cloud agents. New customers buy it on The X Subscription: Free, Plus at $8, Premium at $30, Super at $100, and Ultra at $200. Paid plans share one weekly pool for Cursor's own models with Grok. Third-party models are not included; on Premium and above they bill at API list price from prepaid credit. Older Pro and Pro+ plans still exist only for people who already subscribed.
Desktop editor, CLI, and cloud agents
Skill weights: No code 1/5 · I can follow a change 4/5 · I already develop 5/5.
Free plan
Free · Free Verified
Free includes limited Agent and Chat, plus Tab completions on Cursor's own models. Cursor does not publish a request count.
Limited Agent and Chat requests, and Tab completions on Cursor models (Grok and Composer). No published request count. Third-party models are not available.
The current free plan on cursor.com/pricing. Legacy Hobby is the same kind of limited trial for older accounts.
Vendor does not publish a fixed count.
Paid floor the ranker counts
Plus · $8/mo Verified
Annual is $80/yr on Cursor's pricing docs.
Base weekly pool for Cursor models (Grok 4.7, Grok 4.6, Grok 4.5, and Composer 2.5), shared with Grok. Fast is off and effort is medium. No Auto, Cloud Agents, third-party models, or your own API keys. The pool size is not published as a prompt count.
The entry paid plan. Cursor recommends Premium, not Plus, for new agent users.
Listed, and not used as a price floor
Checked 2026-10-19
A terminal coding agent that shares your Claude plan.
Claude Code is included with paid Claude plans and draws from the same usage pool as Claude on the web, desktop, and mobile. There is no separate Claude Code subscription. The free Claude plan does not include it.
Terminal agent, plus the Claude apps that share one usage pool
Skill weights: No code 0/5 · I can follow a change 3/5 · I already develop 5/5.
Free plan
Free · Free Verified
Free Claude does not include Claude Code. You need Pro or Max before this tool exists for you.
Claude chat only. Claude Code is not included.
Do not plan a coding workflow on the free plan.
Paid floor the ranker counts
Pro · $20/mo Verified
$17/mo when billed annually ($200 up front).
Includes Claude Code. At least 5x Free chat usage per five-hour session, plus weekly limits. Shared with claude.ai. No fixed prompt count.
Anthropic's pricing page lists $20 monthly and $17/mo annual.
Checked 2026-10-19
A coding agent included with ChatGPT, not a separate app subscription.
Codex is the coding agent in the ChatGPT apps and CLI. You do not buy it on its own. Free and Go include a limited trial. Plus is the first plan with expanded Codex. Pro tiers are for longer daily sessions, and credits extend eligible use after the included allowance.
Desktop app, CLI, and IDE, included with ChatGPT
Skill weights: No code 1/5 · I can follow a change 4/5 · I already develop 5/5.
Free plan
Free · Free Verified
Free includes a limited Codex trial. OpenAI does not publish a fixed prompt count. It is enough to see the product, not to build through the month.
Limited trial access to Codex and limited access to GPT-5.6. Codex pricing docs also list GPT-6 Luna at Standard speed in the desktop app, subject to rollout. No published prompt count.
Confirmed on the Codex pricing page as a limited trial. GPT-6 Luna on Free is listed as subject to rollout, so it is not treated as a guarantee.
Vendor does not publish a fixed count.
Paid floor the ranker counts
Plus · $20/mo Verified
Expanded Codex for a few focused sessions a week. Codex app on macOS and Windows. Local and cloud usage share the allowance. Weekly limits may apply. GPT-6 Astra access is limited, with credits for more.
Plus is $20/mo on OpenAI's help center. Annual billing is not offered.
Listed, and not used as a price floor
Codex on the expanded plan is described for a few focused sessions a week. Local and cloud usage share that allowance. That is a match if coding is not the whole workday.
Cursor Premium is the plan the vendor points new agent users toward. If the weekly pool is empty midweek, move up a tier instead of relying on top-ups. Claude Code's pressure shows up as a five-hour window plus a weekly cap, shared with chat.
Use the free trial that matches the shape. Cursor Free to see the editor. Codex Free to see the agent inside ChatGPT. Antigravity if you want a weekly editor-agent quota at $0. Do not use free Claude as a stand-in for Claude Code.
When a pool hits zero, follow the limit playbook for that tool before you switch brands. Two-tool pages: Cursor vs Claude Code, Claude Code vs Codex, Cursor vs Codex.
Kiro, Amp, and Jules are the other repo agents in the catalog. They are not a fourth, fifth, and sixth version of the three above. Each one has a different meter.
Pair pages: Cursor vs Kiro, Amp vs Cursor, Jules vs Codex.
Not if the first project is a landing page or a no-code app. Those three assume you can look at a repo. Cursor is the interactive editor. The cheapest plan the ranker counts is Plus at $8/mo, and it is limited to Cursor's own models. Premium is the tier the vendor recommends for new agent users. Claude Code is a terminal agent with no separate subscription, and the free Claude plan does not include it. Codex comes with ChatGPT. Plus at $20/mo is the first expanded Codex plan. If you want to try an editor agent at $0 first, Google Antigravity is the free individual plan in this catalog.
Free Claude does not include Claude Code. You need Pro or Max before this tool exists for you. The paid floor the ranker counts is Pro at $20/mo. Chat and Claude Code share one usage pool.
Go is listed as "Confirm on chatgpt.com/pricing". Left out of budget matching. The $8 figure is on a Codex pricing doc, and the live ChatGPT pricing page still does not show it. It does not count as a budget floor until a dollar amount is on the live pricing page. Plus at $20/mo is the first plan that does.