OpenAI Codex vs Claude Code vs GitHub Copilot vs Cursor

Pick up to 4 tools and see their free plans, entry prices, platforms and limits in one table. Every fact comes from each vendor's own pages, checked 27 October 2026.

Written head to head: Cursor vs GitHub Copilot · Claude Code vs OpenAI Codex · GitHub Copilot vs OpenAI Codex · GitHub Copilot vs Claude Code, with when to choose each side.

Compared on	OpenAI Codex Free plan · from $8/mo Visit Remove	Claude Code From $17/mo Visit Remove	GitHub Copilot Free plan · from $10/mo Visit Remove	Cursor Free plan · from $20/mo Visit Remove
What it is	OpenAI's coding agent for the terminal, IDEs, desktop app and cloud	Anthropic's agentic coding tool for the terminal, IDEs, desktop and web	GitHub's AI coding assistant for code completion, chat, agents and review	AI code editor with agents that write and change code
Free plan	ChatGPT Free includes limited Codex use for quick coding tasks; Free and Go use GPT-6 Luna at standard speed in the desktop app, subject to rollout	No	2,000 code completions per month and a limited monthly allowance of AI credits for chat and agent features; models by auto selection; agent mode and Copilot CLI included; no cloud agent or code review; no credit card required	Hobby plan: limited Agent requests, access to Composer, no credit card required
Cheapest paid plan	$8/mo (ChatGPT Go)OpenAI Codex pricing	$17/mo billed annually (Claude Pro)Claude Code pricing	$10/mo (Pro)GitHub Copilot pricing	$20/mo (Pro)Cursor pricing
What it covers	Codex comes with ChatGPT plans; Plus ($20/month) expands Codex usage. Codex can also be used with an API key at standard API rates.	Claude Code is included in Pro, Max, Team and Enterprise and shares usage limits with the subscription. Annual Pro billing is $200 up front. Console accounts pay standard API rates instead.	Unlimited code completions and next edit suggestions; 1,500 AI credits a month (1,000 base and 500 flex).	The page says yearly billing saves 20%; yearly per-month prices were not shown.
Free trial	None stated	None stated	None stated	None stated
Platforms	macOS, Windows, Linux, Web, API	macOS, Windows, Linux, Web, iOS, Android, API	Web, macOS, Windows, Linux	macOS, Windows, Linux, Web
Integrations (vendor's figure)	Not stated	Not stated	Not stated	Not stated
Open source	Yes	No	No	No
Good for	Building a small feature from a plain-language description Reviewing code changes before merging Investigating and fixing failing builds	Fixing a bug by pasting in the error message Writing tests and updating dependencies across a project Automating code review on GitHub	Getting code suggestions while writing scripts or apps Asking questions about an unfamiliar codebase Reviewing pull requests before merging	Building or changing a small app by describing the change Fixing bugs across several files Running coding tasks in the cloud while away from the desk
Watch out for	Usage limits depend on the ChatGPT plan, the model and task size, measured per five-hour window Free and Go include only limited Codex access API-key use is billed separately at API rates	Not included in the free Claude plan Usage counts against the same limits as the Claude subscription The desktop app requires a paid subscription	The Free plan is capped at 2,000 completions per month Paid plans include a monthly AI credit allowance (1,500 on Pro); extra usage costs $0.01 per credit Cloud agent and code review are not in the Free plan	The Hobby plan has limited Agent requests Bugbot is billed on usage on top of the plan price Enterprise pricing requires contacting sales
Made by	OpenAI	Anthropic	GitHub (Microsoft)	Anysphere, Inc.

Entry prices are the cheapest paid plan, as each vendor publishes it; “What it covers” says what that price is per (a user, a channel, a brand, a number of contacts). The sources are listed on each tool's page.

Every tool in AI coding assistants is on its category page, with the written head-to-heads on Compare.