OpenAI list rates

Codex Cost Calculator

Estimate Codex token cost from input, output, and cached input. The same page compares that API-equivalent total with ChatGPT Go, Plus, and Pro. No OpenAI key is required.

Claude Code Codex Cursor

Model

Default coding model called out for ChatGPT Plus and Pro Codex sign-in.

Showing an example: One Codex turn on GPT-6.1 Sol. Edit a field, or clear it, to price your own usage.

Uncached input tokens

$2.00 / MTok

Output tokens

$10.00 / MTok

Cached input

$0.10 / MTok

30-minute cache writes

$2.50 / MTok

12-hour cache writes

$5.00 / MTok

Repeat this workload

Scales the dollar total after long-context pricing is chosen, so 20 short requests stay on the short-context rate.

Adjustments

Prompt length pricing

Above 272K input tokens, the long-context card applies to the whole request.

Fast modeFast mode is 2× the standard token rates on this model.Regional / US routing (×1.1)Multiplies every token rate. Off by default.

API-equivalent cost

$1.00

Checked November 2, 2026StandardExample

GPT-6.1 Sol. List rates only. This is not an invoice.

Cost by token type
Type	Tokens	Cost
Uncached input$2.00 / MTok	100,000	$0.20
Cached input$0.10 / MTok	100,000	$0.01
30-minute cache writes$2.50 / MTok	0	$0.00
12-hour cache writes$5.00 / MTok	38,000	$0.19
Output$10.00 / MTok	60,000	$0.60

Against plan stickers

ChatGPT Go
$8.00/mo
$1.00 is 13% of the $8.00 ChatGPT Go sticker
13% of the monthly sticker
Go does not publish a token allowance. Codex on this plan is the limited tier.
ChatGPT Plus
$20.00/mo
$1.00 is 5.0% of the $20.00 ChatGPT Plus sticker
5.0% of the monthly sticker
Plus limits are not a token count. Credits can extend usage and are not included in this sticker.
ChatGPT Pro 100
$100.00/mo
$1.00 is 1.0% of the $100.00 ChatGPT Pro 100 sticker
1.0% of the monthly sticker
The sticker is not an API credit. Usage still depends on the model and the task.
ChatGPT Pro 200
$200.00/mo
$1.00 is 0.50% of the $200.00 ChatGPT Pro 200 sticker
0.50% of the monthly sticker
Pro is billed monthly. The allowance is still not published as a token quota.
ChatGPT Pro 500
$500.00/mo
$1.00 is 0.20% of the $500.00 ChatGPT Pro 500 sticker
0.20% of the monthly sticker
Ultrafast token rates are higher than standard. Turn on Fast mode to estimate that tier.

Assumptions behind the comparison

Subscription plans do not publish a token quota. Session and weekly limits depend on model, prompt length, files, and tools.
The percentage next to a plan is API-equivalent cost divided by the monthly sticker. It is not remaining included usage.
Max and Ultra include monthly API credits that Anthropic says cannot be used in Claude Code or the Claude apps. Those credits are not subtracted.
Claude Pro annual billing is $200 up front (about $17 a month). Comparisons use the $20 monthly sticker.
ChatGPT Pro is monthly only, at $100, $200, or $500. Codex shares usage with the rest of the plan.
Cursor Plus through Ultra include a weekly pool for Cursor models (Composer and Grok), shared with Grok apps. The pool is not a dollar credit equal to the subscription price.
On Cursor Premium and above, third-party models are billed from the first request at API list price, paid from prepaid credit. That spend is not included in the subscription sticker.
The Cursor Token Rate is $0.25 per million tokens on Teams and Enterprise third-party requests. Individual plans do not pay it, and Cursor's own models are exempt. It is off unless you turn it on.
Haiku 5.5's higher rates apply per request whose prompt exceeds 100,000 tokens, including cache reads and cache writes. A summed month of short requests should stay on the standard card. Use Standard rates, or scale with Repeat, so the total is not treated as one giant prompt.
OpenAI long-context rates apply to the whole request when input-side tokens exceed 272,000. Grok 4.7 long context (above 256,000) is 2×, and fast long context is 3×, per Cursor's card.
When a log has a flat cache-write total and no 5-minute / 1-hour split, writes are priced at the cheaper 5-minute rate. Claude Code often uses the 1-hour cache, which costs more.
On Codex logs, cached input is treated as a subset of input tokens. On Claude logs, input tokens exclude cache reads and cache writes.

Prices are approximate public list rates. Anthropic, OpenAI, and Cursor change them without notice. This tool does not call an API. It leaves out tax, negotiated discounts, prepaid credits, batch and flex tiers, and per-call tool fees. A plan comparison is sticker price versus API-equivalent token cost. It is not a promise that the workload fits inside the plan.

How the Codex cost calculator counts tokens

Session files often contain a running total_token_usage. Summing every snapshot would count the session many times. This calculator keeps the largest cumulative total and ignores per-turn lines in the same file. A codex exec --json stream of turn.completed events, with no cumulative total, is summed instead.

One Codex turn on GPT-6.1 SolGPT-6.1 Sol · Standard · total $1.00

Token type	Tokens	Rate	Cost
Uncached input	100,000	$2.00 / MTok	$0.20
Cached input	100,000	$0.10 / MTok	$0.01
12-hour cache writes	38,000	$5.00 / MTok	$0.19
Output	60,000	$10.00 / MTok	$0.60

Prompts over 272,000 input-side tokens, including cached input, use the long-context card for the whole request on models that publish one. The example turn above stays under that line. GPT-5.6 Sol's list price is marked promotional through at least November 21, 2026.

Codex cached input versus Claude cache reads

On Codex logs, cached_input_tokens is treated as part of input_tokens, not as an extra pile. Uncached input is the difference. Reasoning tokens are treated as already included in output unless they are larger than the output count.

That is the opposite of a Claude Code log, where input_tokens excludes cache reads and cache writes. Using the Claude split on a Codex file would double-count the cached prompt.

Codex cost versus ChatGPT Plus and Pro

Go is $8, Plus is $20, and Pro is $100, $200, or $500 a month. Codex is included, and it shares usage with ChatGPT. None of those plans publish a token allowance you can subtract from. The cards under the total only compare API list price with the sticker. Credits you buy on top are extra.

For Claude Code sessions, use the Claude Code cost calculator. For Composer and Grok, use the Cursor cost calculator. For a side-by-side of the three, use the coding agent cost calculator.