WhichAIPlan / Compare / Cursor vs Claude Code vs Codex
checked 2026-10-14
Cursor vs Claude Code vs Codex
Plans, prices and usage rules for Cursor, Claude Code and Codex, copied from each vendor's pricing page on 2026-10-14. Where a vendor doesn't publish a number, we say unknown.
At a glance
| Cursor | Claude Code | OpenAI Codex | |
|---|---|---|---|
| Vendor | Anysphere | Anthropic | OpenAI |
| Runs in | Cursor (VS Code fork) | Terminal / any editor | CLI, IDE extension, web, Codex app |
| Cheapest paid plan | Plus $8/mo | Pro $20/mo | Go $8/mo |
| How limits work | Shared AI usage pool across Cursor and Grok; allowance refreshes weekly; third-party models billed separately at API list price; top-offs available. | Rolling 5-hour session window plus weekly limits on paid plans; Claude Code shares the pool with Claude chat; no fixed message count published; optional usage credits at API rates. | Included in ChatGPT plans. Plus: estimated local messages per 5-hour window (e.g. GPT-6.1 Sol 15-160, GPT-6 Astra 5-45), weekly limits may apply. Pro: no 5-hour limit currently. Credits extend usage. |
| Official pricing | cursor.com/pricing | claude.com/pricing | chatgpt.com/codex/pricing/ |
Entry price difference
Cheapest paid plans differ by $12/mo (2.5x). Entry prices only; usage allowances differ, see the tables.
Cursor: Conflict on 2026-10-14: cursor.com/pricing lists Free/Plus $8/Premium $30/Super $100/Ultra $200 (bundled with X/Grok), while cursor.com/docs/models-and-pricing still lists Pro $20/Pro Plus $60/Ultra $200. We show the pricing page and flag the docs figures.
Every plan
| Tool | Plan | Price | Usage / limits | Source |
|---|---|---|---|---|
| Cursor | Free | $0 | Limited agent and chat requests | source 2026-10-14 |
| Cursor | Plus | $8/mo | Basic shared pool; Grok models only | source 2026-10-14 |
| Cursor | Premium | $30/mo | 4x Plus; third-party models at API list price; cloud agents 4× Plus | source 2026-10-14 |
| Cursor | Super | $100/mo | 15x Plus; priority access at peak 15× Plus | source 2026-10-14 |
| Cursor | Ultra | $200/mo | 50x Plus 50× Plus | source 2026-10-14 |
| Cursor | Pro (per docs page) conflict | $20/mo | Two pools: Cursor models + third-party at API price. Docs: daily agent users typically $60-100/mo total | source 2026-10-14 |
| Claude Code | Free | $0 | No Claude Code | source 2026-10-14 |
| Claude Code | Pro | $20/mo $17/mo annual | At least 5x Free per 5-hour session; exact tokens unknown 5× Free | source 2026-10-14 |
| Claude Code | Max 5x | $100/mo | 5x Pro per 5-hour session; $100/mo API credits 5× Pro | source 2026-10-14 |
| Claude Code | Max 20x | price unknown | 20x Pro per 5-hour session; $200/mo API credits. Price not shown on card ('From $100'); third-party sources say $200 - unverified 20× Pro | source 2026-10-14 |
| OpenAI Codex | Free | $0 | Limited Codex | source 2026-10-14 |
| OpenAI Codex | Go | $8/mo | Limited Codex; may include ads | source 2026-10-14 |
| OpenAI Codex | Plus | $20/mo | Expanded Codex; 5-hour message estimates by model | source 2026-10-14 |
| OpenAI Codex | Pro 100 | $100/mo | No 5-hour limit currently; exact allowance unknown | source 2026-10-14 |
| OpenAI Codex | Pro 200 | $200/mo | More than Pro 100; new subs get lower allowance than before (grandfathering to 2026-10-29) | source 2026-10-14 |
| OpenAI Codex | Pro 500 | $500/mo | Highest allowance; Ultrafast mode | source 2026-10-14 |
Quick read
- Cursor: runs in Cursor (VS Code fork); paid plans from $8/mo. Cursor plans
- Claude Code: runs in Terminal / any editor; paid plans from $20/mo. Claude Code plans
- OpenAI Codex: runs in CLI, IDE extension, web, Codex app; paid plans from $8/mo. Codex plans
Not sure? Answer 4 questions in the plan finder or see the full comparison of all tools.
FAQ
Which is cheaper, Cursor, Claude Code or Codex?
Cursor's cheapest paid plan is Plus at $8/mo (Basic shared pool; Grok models only); Claude Code's cheapest paid plan is Pro at $20/mo (At least 5x Free per 5-hour session; exact tokens unknown); OpenAI Codex's cheapest paid plan is Go at $8/mo (Limited Codex; may include ads). Source: official vendor pricing pages, verified 2026-10-14.
How do the usage limits differ?
Cursor: Shared AI usage pool across Cursor and Grok; allowance refreshes weekly; third-party models billed separately at API list price; top-offs available. Claude Code: Rolling 5-hour session window plus weekly limits on paid plans; Claude Code shares the pool with Claude chat; no fixed message count published; optional usage credits at API rates. OpenAI Codex: Included in ChatGPT plans. Plus: estimated local messages per 5-hour window (e.g. GPT-6.1 Sol 15-160, GPT-6 Astra 5-45), weekly limits may apply. Pro: no 5-hour limit currently. Credits extend usage.
Where does each tool run?
Cursor runs in Cursor (VS Code fork); Claude Code runs in Terminal / any editor; OpenAI Codex runs in CLI, IDE extension, web, Codex app.
What does each plan cost?
Cursor: Free $0, Plus $8/mo, Premium $30/mo, Super $100/mo, Ultra $200/mo; Claude Code: Free $0, Pro $20/mo, Max 5x $100/mo, Max 20x price unknown; OpenAI Codex: Free $0, Go $8/mo, Plus $20/mo, Pro 100 $100/mo, Pro 200 $200/mo, Pro 500 $500/mo.
Head-to-head comparisons
Get the weekly what-changed email
One email a week listing the price and limit changes we logged.