OpenAI model cost
GPT-5.2 / 5.2 Codex cost in Cursor vs Claude Code per 1M tokens
GPT-5.2 / 5.2 Codex costs $1.75 per million input tokens and $14 per million output tokens, with cache reads at $0.175. An hour of agent coding runs about $1.12 at list price.
Prices last checked 2026-11-09. Independent comparison; not affiliated with any vendor named.
| Per 1M tokens | Input | Cache write | Cache read | Output |
|---|---|---|---|---|
| GPT-5.2 / 5.2 Codex | $1.75 | $1.75 | $0.175 | $14 |
A typical session
A 30-minute agent session reads about 1M tokens from cache, writes 60k to cache, sends 20k fresh input and gets 17.5k back. On GPT-5.2 / 5.2 Codex that is about $0.56.
In Cursor
In Cursor it is a third-party model: on Premium, Super and Ultra it is billed at these list prices from prepaid credit from the first request. It is not available on Plus.
In Claude Code
Claude Code runs Anthropic models only, so GPT-5.2 / 5.2 Codex is not an option there.
Quick calculator
EstimateAbout $1.20 per active hour, so $101/mo at list price.
- Pay-as-you-go API Cheapest $101
- Cursor Plus Can't run this model$109
- Claude Pro Can't run this model$121
- Cursor Premium $131
- Cursor Super $201
- Claude Max Can't run this model$201
- Cursor Ultra $301
- Claude Ultra Can't run this model$301
Assumes about 2.2M tokens per active hour, mostly cache reads, which is typical for agent sessions. Your real mix can be very different: drop your own export for exact numbers.
FAQ
How much does GPT-5.2 / 5.2 Codex cost per 1M tokens?
$1.75 input, $1.75 cache write, $0.175 cache read and $14 output, per million tokens (source: Lab list price).
Is GPT-5.2 / 5.2 Codex included in Cursor plans?
In Cursor it is a third-party model: on Premium, Super and Ultra it is billed at these list prices from prepaid credit from the first request. It is not available on Plus.
Can I use GPT-5.2 / 5.2 Codex in Claude Code?
Claude Code runs Anthropic models only, so GPT-5.2 / 5.2 Codex is not an option there.
Related
What does your AI coding cost?
Drop your Cursor CSV or ccusage output. It's read in your browser and takes under a minute.
Analyze my usage free