Compare
Codex Pro vs Claude Code Max (2026): $100 vs $100, $200 vs $200
Codex Pro vs Claude Code Max, tier for tier: $100 and $200 plans side by side, what each limit resets on, and when credits or a cheaper plan is enough.
7 min read
Prices and limits checked Oct 18–25, 2026 on OpenAI’s and Anthropic’s own pages; each date is under Sources. No affiliate links on this page (how we work).
OpenAI and Anthropic both landed on $100 and $200 for their heavy coding plans. The price tags match. What you get for them doesn’t, and only one of the two companies tells you how big its tiers are.
Short version: Claude Max tells you its multiple (5x Pro on Max, 10x Pro weekly on Ultra) and its reset rules. Codex Pro tells you there’s “currently” no five-hour limit and little else. This page is only the paid-tier matchup; for features and workflow, read Codex vs Claude Code.
Quick answer, tier by tier
- $100: Codex Pro $100 vs Claude Max 5x. Max 5x gets “five times the Pro plan’s per-session usage allowance” (verified Oct 24, 2026). Codex Pro $100 gets no five-hour limit, but OpenAI doesn’t say how much usage it includes (verified Oct 25, 2026).
- $200: Codex Pro $200 vs Claude Ultra (formerly Max 20x; claude.com/pricing, verified Nov 1, 2026). Ultra is 10 times Pro every week with no five-hour limit. Codex Pro $200 “includes more usage than Pro 100”, with no number attached, and new subscribers get “a lower usage allowance” than people who were grandfathered in (verified Oct 21, 2026).
- $500: Codex Pro $500 only. It’s the one tier with Ultrafast mode. Anthropic doesn’t sell anything above Ultra for individuals.
The price table
| Tier | Monthly | Billing | Overflow |
|---|---|---|---|
| Codex Pro $100 | $100 | Monthly only | Credits, API key |
| Claude Max 5x | $100 | Monthly only | Usage credits |
| Codex Pro $200 | $200 | Monthly only | Credits, API key |
| Claude Ultra | $200 | Monthly only | Usage credits |
| Codex Pro $500 | $500 | Monthly only | Credits, API key |
USD. Codex: Codex pricing (verified Oct 25, 2026); no annual billing per OpenAI’s Pro plans article (verified Oct 21, 2026). Claude: Max plan article, “monthly subscription only”, web prices; “Mobile pricing may vary” (verified Oct 24, 2026).
$100 tier: Codex Pro $100 vs Claude Max 5x
Claude Max 5x gives you five times Pro’s per-session allowance, with sessions resetting every five hours and a weekly limit on top. Claude chat and Claude Code share that pool, so an afternoon research chat eats into your evening coding (Claude Code with Pro or Max, verified Oct 23, 2026). Max also comes with $100 a month in API credits, but those “can’t be used in Claude Code or the Claude apps” and don’t roll over (verified Oct 18 and Oct 23, 2026). They’re for your own scripts, not your terminal agent.
Codex Pro $100 removes the five-hour window: “Pro plans currently have no five-hour limit”. OpenAI adds “Weekly limits may also apply” and points you to your usage dashboard for “current limits and reset times” (verified Oct 25, 2026). Plus gets published estimates (15–160 local messages per five hours on GPT-6.1 Sol); Pro gets nothing comparable, and no Ultrafast. So at $100, Claude sells a known multiple, and OpenAI sells the absence of one limit.
$200 tier: Codex Pro $200 vs Claude Ultra
Claude Ultra is 10 times Pro every week with no five-hour limit, plus $200 a month in API credits under the same rules as above. Max plans also have “a separate weekly limit for Fable”, and Anthropic doesn’t publish a different Fable figure for 5x and 20x (verified Oct 23–24, 2026). If you’re moving up from 5x mid-month, the upgrade is “charged on a prorated basis for the remainder of your billing cycle” (verified Oct 24, 2026). The full 5x-vs-20x breakdown is in Claude Max vs Ultra.
Codex Pro $200 “includes more usage than Pro 100”. How much more isn’t published. Then there’s the part that annoys me: people on Pro 200 who were subscribed Sep 22–29, 2026 keep the old allowance “through October 29, 2026”, and new subscriptions that aren’t eligible get “a lower usage allowance” (verified Oct 21, 2026). Same $200, less usage if you’re new, and no number either way. Ultrafast still isn’t included, and “At launch, buying credits on Pro 100 or Pro 200 does not unlock Ultrafast.”
Two heavyweights at $200, Alien vs. Predator rules: whoever wins, your card loses. Only one fighter has published its weight.
Codex Pro $500 has the highest included usage of the three Pro tiers and is the only one with Ultrafast, which burns included usage at 8x the Standard rate (verified Oct 25, 2026). Claude has nothing to compare it with.
Limits and resets, side by side
| Rule | Codex Pro | Claude Max |
|---|---|---|
| 5-hour window | None “currently” | Yes, resets every 5h |
| Weekly limit | “May also apply” | Yes, fixed reset time |
| Published size | Not published | 5x Pro (Max), 10x weekly (Ultra) |
| Shared pool | Codex + ChatGPT Work | Claude chat + Claude Code |
Sources: Codex pricing (verified Oct 25, 2026); Max plan article (verified Oct 24, 2026). Claude’s weekly reset “resets at a fixed time each week that is assigned to your account” and shows in Settings > Usage (verified Oct 23, 2026).
OpenAI charges up to $500 a month, and its answer to “how much do I get?” is “check your dashboard”. Logan Roy would read that page and say “You are not serious people.” Publish the multiple.
When you hit the cap
On Codex Pro, Plus and Pro users “can purchase additional credits to continue working without needing to upgrade” (verified Oct 25, 2026). The per-credit price isn’t public: “Credit purchase prices and applicable discounts depend on your plan or agreement.” Credits are valid for 12 months from purchase (verified Oct 24, 2026). Or run extra local chats on an API key at API rates. More in Codex credits.
On Claude Max, your options at the limit are to move from Max to Ultra, turn on usage credits, switch to a Console account with API credits, or wait for the reset (verified Oct 23, 2026). Usage credits are billed “at standard API pricing rates”, and “In most cases, usage credits do not expire” (verified Oct 21, 2026). So Claude’s overflow has a published price; Codex’s shows up at checkout.
Which to pick
We haven’t run either plan, so this goes by what each company publishes:
- You want to know what you’re paying for: Claude Max. 5x Pro on Max, 10x Pro weekly on Ultra, a five-hour reset on Max and a fixed weekly reset are all in writing.
- The five-hour window is what keeps stopping you: Codex Pro, which “currently” has no five-hour limit. Watch the weekly one.
- You call models from your own scripts too: Claude Max, because $100 or $200 of monthly API credits come with it (not usable in Claude Code).
- You need the fastest mode either side sells: Codex Pro $500, the only tier with Ultrafast.
- You only overflow a few times a month: stay on the $20 plan and pay for overflow instead. See Codex Pro vs Plus for the OpenAI side.
Other tools at these prices are on AI coding tools compared.
Here’s what I’d do
If my agent is already picked, that decides it, and I’d buy the $100 tier first. If I’m torn, Claude Max 5x, purely because I can read what I’m getting. New Codex Pro $200 buyers get the lower allowance, so I’d prove $100 isn’t enough first. Move up after a month of hitting limits, not one bad Tuesday.
This page compares the coding agents. To compare the whole subscriptions (chat, images, extras), see Claude Max vs ChatGPT Pro.
FAQ
Is Codex Pro or Claude Max better value? Neither publishes enough to say per dollar; Claude Max states 5x Pro (Max) or 10x Pro weekly (Ultra) usage, while OpenAI publishes no Pro usage figures.
Does Claude Max include Claude Code? Yes, Claude Code is on all paid Claude plans and shares one usage pool with Claude chat; Max’s monthly API credits can’t be used in Claude Code.
Can I buy more Codex usage without Pro? Yes, Plus users who reach their limit can buy ChatGPT credits, and anyone can run extra local chats with an API key at API rates.
Is Codex Pro $200 the same as Claude Ultra? Same price, different terms: Ultra is 10 times Pro every week with no five-hour limit, while OpenAI only says Pro 200 includes more usage than Pro 100.
Sources
- OpenAI, Codex pricing (plans, Pro five-hour note, weekly limits, speed multipliers, credits, limit behaviour): https://developers.openai.com/codex/pricing (verified 2026-10-25)
- OpenAI Help Center, About ChatGPT Pro plans (Pro 100/200/500 usage wording, Ultrafast, Pro 200 grandfathering, monthly billing): https://help.openai.com/en/articles/9793128-about-chatgpt-pro-plans (verified 2026-10-21)
- OpenAI Help Center, Using credits for flexible usage (12-month validity): https://help.openai.com/en/articles/12642688-using-credits-for-flexible-usage-in-chatgpt-personal-plans (verified 2026-10-24)
- Anthropic, What is the Max plan? (prices, multiples, resets, monthly only, proration): https://support.claude.com/en/articles/11049741-what-is-the-max-plan (verified 2026-10-24)
- Anthropic, Use Claude Code with your Pro or Max plan (shared limits, Fable limit, options at limit, wrap-up, API credits): https://support.claude.com/en/articles/11145838-use-claude-code-with-your-pro-or-max-plan (verified 2026-10-23)
- Anthropic, Usage credits for paid Claude plans: https://support.claude.com/en/articles/12429409-extra-usage-for-paid-claude-plans (verified 2026-10-21)
- Anthropic, Monthly API credits for Max and Team plans: https://support.claude.com/en/articles/17154008-monthly-api-credits-for-max-and-team-plans (verified 2026-10-18)