Skip to content
Estimates onlyCode Cost Calculator

Prices checked

Claude Code, Codex, and Cursor

Coding Agent Cost Calculator

Compare API-equivalent cost for Claude Code, Codex, and Cursor using the public list rates on this site. The figures below are estimates, not invoices, and plan stickers are not token balances.

Claude Code vs Codex vs Cursor on the same tokens

This row uses one Claude Code-shaped day: 85,000 uncached input tokens, 220,000 output tokens, 18 million cache reads, and 1.2 million 1-hour cache writes. Each column prices that same mix on the default model for that agent. Codex crosses its 272,000-token long-context line because cache reads count toward the prompt. Cursor has no separate cache-write price, so those writes are counted at the input rate.

Same token mix priced for Claude Code, Codex, and Cursor
AgentModelRate cardAPI-equivalent
Claude CodeClaude Sonnet 5.5Standard$8.97
CodexGPT-6.1 SolLong context$19.24
CursorComposer 2.5Standard$4.79

A realistic session for each coding agent

The shared mix above is one prompt shape. Real sessions look different. These are the worked examples from each calculator, still at list price.

Claude Code cost for a heavy day

One heavy Claude Code day on Sonnet 5.5 on Claude Sonnet 5.5 is $8.97. That sits next to Claude Pro at $20 a month on the Claude Code cost calculator, which does not mean the day fits inside Pro.

Codex cost for one turn

One Codex turn on GPT-6.1 Sol is $1.00 at the standard card. ChatGPT Plus is $20 a month and does not publish a token allowance. Price a Codex log on the Codex cost calculator.

Cursor cost for a Composer session

One Cursor agent session on Composer 2.5 is $0.75 at Cursor's published overage rates. Cursor Plus is $8 a month for Cursor models only. Third-party models are extra. Use the Cursor cost calculator for a Composer, Grok, or list-price model.

Plan stickers are not token quotas

Claude Pro, ChatGPT Plus, and Cursor Plus all cost a monthly fee and all limit usage without publishing that limit as a token count. Max, Ultra, and higher Cursor tiers raise the window or the pool. They do not hand you a dollar credit equal to the sticker. The comparison on each calculator is API-equivalent cost divided by that monthly price.

  • Claude Code: Claude Pro at $20.00 a month. No published token quota. Pro has a rolling five-hour limit and a weekly limit. This $20 is the monthly sticker, not a credit balance.
  • Codex: ChatGPT Plus at $20.00 a month. Plus limits are not a token count. Credits can extend usage and are not included in this sticker.
  • Cursor: Cursor Plus at $8.00 a month. Third-party models are not on Plus. The weekly pool for Composer and Grok is shared with Grok and is not a token quota.

Which coding agent cost calculator to open

Use the page that matches the log you have. The math stays in the browser.