For a month, ChatGPT Pro 200 looked like a plan to leave. OpenAI's 2026-09-29 email told subscribers that the $200 tier would move to a lower Work and Codex allowance from 2026-10-30, and a lot of buying advice, including earlier articles on this site, was written around that cut. On 2026-10-29, one day before it would have taken effect, OpenAI canceled it. The release notes entry of that day says Pro 100 and Pro 500 now include twice their previous Work and Codex allowance, and Pro 200 subscribers who had been moved to the lower allowance were doubled back up to the higher one. Anything you read that still says Pro 200 runs on a reduced allowance from October 30 is out of date.

OpenAI Help Center: About ChatGPT Pro tiers · OpenAI Help Center: ChatGPT release notes, entry of 2026-10-29

That reversal changes the arithmetic of every ChatGPT buying decision, which is why this article is a planner rather than a review. It walks four usage profiles, chat only, daily Codex, agents all day, and Ultrafast, through what OpenAI actually publishes (a credit rate card and five-hour message estimates, not Pro allowance numbers), then ends each profile with one product to buy on this site: a top-up on your own account where your account is eligible, a ready-made account where it is not. All vendor facts were re-read on 2026-11-10; all prices in the tables below are live and never typed into the text.

1. What changed on 2026-10-29, in OpenAI's own words

The help center article About ChatGPT Pro tiers, updated around 2026-11-03, now carries four statements that matter for buyers. First, Pro 200 is available for new subscriptions again at $200 per month. Second, the reduction announced in the September 29 email is canceled: subscribers who already had the higher Work and Codex allowance keep it, and those who had the lower one saw it doubled to match. Third, Pro 100 and Pro 500 include twice their previous Work and Codex allowance, applied automatically to existing subscribers without waiting for renewal. Fourth, Ultrafast remains available only on Pro 500, and buying credits on Pro 100 or Pro 200 does not unlock it.

OpenAI Help Center: About ChatGPT Pro tiers

The chat side moved less. The weekly GPT-6 Pro message limits in Chat for Pro 100 and Pro 500 are unchanged; on Pro 200 the lower weekly limit was raised to match the higher one. OpenAI is explicit that these chat message limits are separate from the Work and Codex allowance, and equally explicit that a larger allowance does not guarantee twice as many completed tasks, because usage depends on model, context, reasoning, tool use and speed mode. Plus usage is unchanged.

OpenAI Help Center: About ChatGPT Pro tiers

The October 29 change by tier, as described by OpenAI's release notes and Pro tiers article (no numeric allowances are published).
TierWork and Codex allowanceWeekly GPT-6 Pro chat limitUltrafast
Pro 100 ($100/mo)Doubled versus before 10-29UnchangedNot included
Pro 200 ($200/mo)Cut canceled; lower allowance doubled to match the higher oneLower limit raised to match the higher oneNot included, even with purchased credits
Pro 500 ($500/mo)Doubled versus before 10-29UnchangedIncluded (GPT-6 Astra and GPT-6.1 Sol)
Plus ($20/mo)UnchangedNot applicable (GPT-6 Pro is a Pro feature)Not included

OpenAI Help Center: About ChatGPT Pro tiers · OpenAI Help Center: ChatGPT release notes, entry of 2026-10-29 · OpenAI: ChatGPT pricing page

Press coverage adds context the help pages omit. gHacks reported on 2026-10-30 that the announcement came from OpenAI's Tibo on X with the words that all limits for Pro 100/200/500 and Business Premium were being doubled after customer feedback, and that extended Dot limits stay separate from the default allowance for now. Treat the quote as a press report; the binding text is the help center.

gHacks: OpenAI doubles ChatGPT Pro and Business Premium usage limits after dropping a planned Pro 200 cut

2. What OpenAI publishes, and what it does not

A spend planner has to be honest about its inputs. OpenAI publishes three kinds of numbers for personal plans. The list prices: Go $8, Plus $20, Pro from $100 with tiers at $100, $200 and $500, all monthly, with no annual Pro billing. The Codex credit rate card: credits per million input, cached-input and output tokens per model, at Standard speed. And the local-message estimates per five-hour window for Plus and Standard Business, by model. It does not publish how many credits, tokens or messages a Pro tier includes per week, before or after the doubling. The Pro tiers article says only that Pro 200 includes more usage than Pro 100 and Pro 500 offers the highest included usage of the three.

OpenAI Help Center: About ChatGPT Pro tiers · Codex documentation: Pricing · OpenAI: ChatGPT pricing page

Two more facts shape the plan. Pro plans currently have no five-hour limit in Codex, so for a Pro subscriber the binding constraint is the weekly Work and Codex allowance, which OpenAI does not quantify. And on Plus or Pro, when you hit a Codex limit you can buy credits if purchases are available to your account; included usage is consumed first, then the credit balance, which is valid for 12 months and does not roll over. Credit pack prices on OpenAI's side vary by account and region and are not listed publicly.

Codex documentation: Pricing · OpenAI Help Center: Using Credits for Flexible Usage in ChatGPT (Personal plans)

Because the Pro allowances are unpublished, this article never tells you that Pro 200 gives you N tasks. It does something more defensible: it uses the published rate card to estimate what a credit pack buys you in tasks, and uses this site's live prices to compute what the gap between two plans buys you in credit packs. Those two computed numbers, plus the five-hour estimates for Plus, are enough to decide.

3. Four usage profiles, and the question each has to answer

Chat only. You talk to ChatGPT, upload files, generate images, and open Codex perhaps once a week to fix a script. The relevant limits are the Plus message and upload limits, which are generous for this pattern, and the Plus five-hour Codex window, which you will rarely touch. Your question is whether you have an account that can hold a Plus subscription at all.

Daily Codex. You run Codex in the CLI or IDE several hours a day on one or two repositories, mostly with GPT-6.1 Sol. On Plus you will meet the five-hour window on busy days, and the weekly limit on busy weeks. Your question is whether buying credit packs on top of Plus stays below the price of Pro 100, and whether you want Pro's absence of a five-hour limit.

Agents all day. You keep cloud tasks running, use ChatGPT Work for long jobs, review pull requests automatically, and want GPT-6 Pro in Chat. Codex and Work share one allowance on Plus and Pro, so your weekly consumption is the sum of everything. Your question is Pro 100 versus Pro 200, and the October 29 doubling of Pro 100 makes that closer than it was.

Ultrafast. You need GPT-6 Astra or GPT-6.1 Sol generating tokens up to 8x faster, in Codex or Work, and you accept that Ultrafast consumes included usage at 8x the Standard rate and purchased credits at 6x. Only Pro 500 has it among self-serve plans; the Codex speed documentation is explicit that other self-serve plans do not get it even with credits. Your question is not which plan but whether speed is worth the Pro 500 price shown in the live table.

OpenAI Help Center: About ChatGPT Pro tiers · Codex documentation: Speed modes and Ultrafast

A fifth situation cuts across all four: you have no ChatGPT account, your card is declined, or your existing account cannot take a subscription. Then the matching ready-made account on this site is the path, and the eligibility table in section 8 tells you which.

4. The official Codex credit rate card

The Codex pricing page quotes Standard-speed rates in credits per million tokens, with separate prices for input, cached input and output, and no separate cache-write charge. The current models are below exactly as read on 2026-11-10. GPT-5.6 Sol is listed at promotional pricing available at least through 2026-11-21; GPT-5.5 left Codex for ChatGPT sign-in on 2026-10-14 and is not in the table.

Codex documentation: Pricing · Codex documentation: Models

Codex credit rates at Standard speed, credits per 1M tokens (learn.chatgpt.com/docs/pricing, read 2026-11-10).
ModelInputCached inputOutputNote
GPT-6 Astra250251,250Pro reasoning model; Ultrafast variant on Pro 500
GPT-6.1 Sol502.5250Default recommendation for Plus and Pro Codex
GPT-6.1 Terra251.25125Rolled out to Plus/Pro/Business on 2026-10-26
GPT-6 Sol505250Previous-generation Sol
GPT-6 Luna2.50.2512.5Light model; also the Free/Go desktop model
GPT-5.6 Sol10010500Promotional pricing at least through 2026-11-21

Codex documentation: Pricing

Speed modes multiply those rates. Fast mode uses included subscription usage at 2.5x and purchased credits at 2x the Standard rate. Ultrafast, for GPT-6 Astra and GPT-6.1 Sol only, uses included usage at 8x and purchased credits at 6x. The practical reading: a credit pack lasts roughly half as long if you leave Fast on, and a Pro 500 subscriber who leaves Ultrafast on burns the included allowance eight times faster than Standard.

Codex documentation: Pricing · Codex documentation: Speed modes and Ultrafast

Speed-mode multipliers on the Standard rate (Codex pricing and speed pages, 2026-11-10).
Speed modeIncluded subscription usagePurchased creditsWho can select it
Standard1x1xAll plans
Fast2.5x2xPlans where available
Ultrafast (GPT-6 Astra, GPT-6.1 Sol)8x6xPro 500 and eligible Enterprise/Edu only

Codex documentation: Pricing · Codex documentation: Speed modes and Ultrafast

OpenAI's own anchor for task size is a single sentence: a typical GPT-5.6 Sol task may use 2 to 15 credits. At the GPT-6.1 Sol rate, which is half the GPT-5.6 Sol promotional rate on every column, a comparable task would land around 1 to 8 credits. The estimator below lets you replace those anchors with your own token counts.

Codex documentation: Pricing

5. Five-hour estimates for Plus, and a credit-cost estimator

For Plus and Standard Business, OpenAI publishes estimated local messages per five-hour window by model. They are estimates, not fixed limits; cloud tasks may use more of the allowance than local messages, and local messages and cloud chats share one allowance with weekly limits possibly applying on top. Pro plans currently have no five-hour limit, which is the single clearest functional difference between Plus and Pro 100 for a Codex user.

Codex documentation: Pricing

Estimated local Codex messages per five-hour window (learn.chatgpt.com/docs/pricing, read 2026-11-10).
ModelPlusStandard BusinessWhat the range means for a daily user
GPT-6 Astra5–455–45Heavy reasoning: a few large tasks can close the window
GPT-6.1 Sol15–16015–160Typical daily coding model; busy afternoons hit the low end
GPT-6.1 Terra35–28535–285Cheaper per message; good for visual and writing tasks
GPT-6 Sol15–15015–150Older Sol; similar budget to 6.1 Sol
GPT-6 Luna350–3,000350–3,000Light model; rarely the constraint

Codex documentation: Pricing

The script below is standard-library Python. It holds the rate card as a dictionary, three illustrative task sizes in tokens, and the purchased-credit speed multipliers, and prints the credit cost per task and how many such tasks a 500-credit pack covers. The task sizes are placeholders; replace them with numbers from your own Codex usage dashboard before trusting any output.

codex_credit_estimator.py — offline; rates from learn.chatgpt.com/docs/pricing as read on 2026-11-10
# codex_credit_estimator.py - offline estimate of Codex credit cost per task.
# Rates: learn.chatgpt.com/docs/pricing, Standard speed, credits per 1M tokens
# (input, cached input, output), read 2026-11-10. Re-check the page before relying on it.
RATES = {
    "gpt-6-astra":   (250, 25, 1250),
    "gpt-6.1-sol":   (50, 2.5, 250),
    "gpt-6.1-terra": (25, 1.25, 125),
    "gpt-6-sol":     (50, 5, 250),
    "gpt-6-luna":    (2.5, 0.25, 12.5),
    "gpt-5.6-sol":   (100, 10, 500),   # promotional rate, at least through 2026-11-21
}
SPEED = {"standard": 1, "fast": 2, "ultrafast": 6}  # multipliers on PURCHASED credits
# Illustrative task sizes in tokens (input, cached input, output). Replace with your own logs.
TASKS = {
    "small fix":      (20_000, 40_000, 3_000),
    "feature branch": (120_000, 300_000, 15_000),
    "repo refactor":  (400_000, 900_000, 40_000),
}
PACK = 500  # credits per unit as sold on this site

def credits(model, tokens, speed="standard"):
    rate_in, rate_cached, rate_out = RATES[model]
    tok_in, tok_cached, tok_out = tokens
    base = (tok_in * rate_in + tok_cached * rate_cached + tok_out * rate_out) / 1_000_000
    return base * SPEED[speed]

if __name__ == "__main__":
    for model in RATES:
        for name, tokens in TASKS.items():
            cost = credits(model, tokens)
            print(f"{model:14s} {name:15s} {cost:8.1f} credits  ~{PACK / cost:7.1f} tasks per {PACK}-credit pack")

Run with the placeholder sizes, the script says a small GPT-6.1 Sol fix costs under two credits and a repository-scale refactor costs around 32, so a 500-credit pack covers roughly 15 large refactors or several hundred small fixes at Standard speed; on GPT-6 Astra the same refactor costs about 170 credits, three per pack. Those are arithmetic consequences of the placeholder token counts and the published rates, not measurements, and they are the reason the model you pick matters more than the plan you pick for credit spend.

6. Live prices on this site, and what the gaps buy

The first table lists the top-up products that go onto your own ChatGPT account. Each row links to the product; the value refreshes from the pricing page, so nothing here goes stale when the catalogue moves. OpenAI's list prices for context are $20 for Plus and $100, $200 and $500 for the three Pro tiers; this site's prices are set independently and are not a discount on those.

OpenAI: ChatGPT pricing page

Top-ups onto your own ChatGPT account (Session precheck first, pay after). Prices follow the pricing page and refresh hourly; the order page is authoritative.
ProductLive priceAccount state accepted
ChatGPT Plus top-up27.58 USDTFree, or Plus already expired
ChatGPT Pro 100 top-up126.68 USDTFree, Plus or Pro; new, renew or upgrade
ChatGPT Pro 200 top-up193.75 USDTFree, Plus or Pro; new, renew or upgrade
ChatGPT Pro 500 top-up506.56 USDTFree, Plus or Pro; the only tier with Ultrafast
Codex credits, 500 per unit (1–20 units per order)26.83 USDTPlus or Pro account; Free accounts cannot top up credits

The second table does the planner's arithmetic on those live values. Each row is a formula over product prices, recomputed whenever a price changes, so the integers below are the current answers and not numbers we typed. Read it with the estimator's output: a 500-credit pack is roughly 15 large GPT-6.1 Sol refactors at the placeholder sizes.

What the price gaps between plans buy in 500-credit packs, computed from the live prices above. Prices follow the pricing page and refresh hourly; the order page is authoritative.
QuestionLive answerHow to read it
Plus plus three credit packs, total108.05 USDTA Plus month with 1,500 purchased credits
Pro 100 minus (Plus plus three packs)18.63 USDTPositive means Plus plus three packs is cheaper this month
Whole packs the Pro 100 minus Plus gap buys3Need more packs than this most months: move to Pro 100
Whole packs the Pro 200 minus Pro 100 gap buys2Pro 100 plus this many packs costs about the same as Pro 200
Whole packs the Pro 500 minus Pro 200 gap buys11Credits never unlock Ultrafast, so this is capacity only
Pro 200 price as a multiple of Pro 1001.53OpenAI's list ratio is 2.00; the tiers' allowances are unpublished

Two readings follow directly. If you buy fewer packs per month than the Pro 100 minus Plus row shows, Plus plus packs is the cheaper way to get the same Codex capacity, with the caveat that Plus keeps its five-hour window and Pro does not. And because the Pro 200 minus Pro 100 gap buys only a small number of packs while Pro 100's allowance was doubled on October 29, Pro 200 is now a purchase for people whose weekly consumption clearly exceeds Pro 100 across Work and Codex combined, or who want the higher Pro 200 weekly GPT-6 Pro chat limit, not a default upgrade.

7. The decision table: profile in, product out

Combine the profiles, the published estimates and the live arithmetic, and the recommendations become short. The table assumes you have a ChatGPT account you can log into; the next section covers the case where you do not.

Recommended purchase per usage profile (this site's catalogue, 2026-11-10).
ProfileSignal in your usageBuyWhy
Chat only, Codex rarelyNever saw a Codex limit messageChatGPT Plus top-upPlus is the first plan with full Codex access; Pro adds nothing you will use
Codex a few times a weekHit the five-hour window one or two days a weekPlus top-up plus 1–2 credit packsPacks cover the overflow for less than the Pro 100 gap; credits valid 12 months
Codex daily, several hoursWindow hit most days, or weekly limit most weeksChatGPT Pro 100 top-upNo five-hour limit; allowance doubled on 10-29; packs would exceed the gap
Agents and Work all dayCloud tasks plus Work plus GPT-6 Pro chat, every dayChatGPT Pro 200 top-upHighest allowance without Ultrafast; GPT-6 Pro chat limit matches the former higher tier
Need Ultrafast or the top allowanceToken speed is the bottleneck, or Pro 200 runs outChatGPT Pro 500 top-upOnly Pro plan with Ultrafast; included usage at 8x, credits at 6x the Standard rate
Two or more people, company purchaseNeed seats and an invoiceChatGPT Business seatsPer seat, minimum two seats; invoice by email for company buyers

OpenAI Help Center: About ChatGPT Pro tiers · Codex documentation: Pricing · OpenAI Help Center: Using Credits for Flexible Usage in ChatGPT (Personal plans)

Three edge cases. If you are on Pro 100 and run out two weeks in three, buy packs for one month and log the count before upgrading; the Pro 200 gap is small, but so is the number of packs it buys, and your log settles it. If you want GPT-6 Pro in Chat at the Pro 200 weekly limit, that alone justifies Pro 200 over Pro 100 plus packs, because credits do not extend chat message limits. And if you were waiting for a cheaper Pro 200 because of the September email, stop waiting: the price is unchanged at $200 on OpenAI's side and the allowance is the higher one.

OpenAI Help Center: About ChatGPT Pro tiers · OpenAI Help Center: Using Credits for Flexible Usage in ChatGPT (Personal plans)

usage_log.csv — one row per week; fill it for a month before changing tier
week_start,plan,codex_limit_hits,hours_blocked,credits_bought,credits_left,cloud_tasks,work_sessions,note
2026-11-03,plus,3,4.5,0,0,2,4,"five-hour window closed Tue and Thu afternoons"
2026-11-10,plus,2,2.0,500,310,3,5,"one 500-credit pack bought 11-11; Fast mode off"
2026-11-17,plus,0,0.0,0,180,1,3,"quiet week; 130 credits used"
2026-11-24,,,,,,,,"decide: packs per month so far = 1; compare with the live gap table"

8. No account, declined card or restricted account: ready-made accounts

A top-up requires an account that OpenAI will let us subscribe: Plus needs a Free account or one whose Plus has expired; Pro 100, 200 and 500 accept Free, Plus or Pro accounts for new subscriptions, renewals and upgrades; credit packs need a Plus or Pro account. If you have no account, cannot pay OpenAI with your card, or your account state does not fit, this site sells ready-made accounts at each tier: one account per order, for one user, with the platform-assigned login mailbox forwarding to your own address. Since 2026-11-09 the ChatGPT Plus ready-made account is delivered already activated.

ClaudeMax FAQ (English)

Ready-made accounts beside the matching top-up, with the premium computed live. Prices follow the pricing page and refresh hourly; the order page is authoritative.
Product or ratioLive valueNote
ChatGPT Plus ready-made account29.66 USDTDelivered ready to use since 2026-11-09
ChatGPT Pro 100 ready-made account133.98 USDTStaff delivery after you enter a receiving mailbox
ChatGPT Pro 200 ready-made account201.05 USDTSame tier as the top-up above
ChatGPT Pro 500 ready-made account506.56 USDTUltrafast on a ready-made account
Plus ready-made divided by Plus top-up1.08Premium for not needing your own eligible account
Pro 200 ready-made divided by Pro 200 top-up1.04Premium at the most popular tier
Pro 200 ready-made minus Pro 200 top-up7.31 USDTThe absolute difference this month

The ratios are small, which is deliberate: a ready-made account is a convenience and eligibility product, not a different plan. What you give up is continuity. A top-up keeps your chat history, memory, projects and custom GPTs on the account you already use; a ready-made account starts empty, and the login mailbox is assigned by the platform and cannot be changed, though mail to it is forwarded to the mailbox you enter on the order page. Our sibling article in this batch compares the two routes across ChatGPT and Claude in detail.

Eligibility and delivery per product on this site (catalogue of 2026-11-10; wording from the site FAQ).
ProductAccount state acceptedDeliveryTypical timing shown on the order page
Plus top-upFree, or expired PlusSession JSON on the order page, precheck, then payUsually 5–30 minutes
Pro 100 / 200 / 500 top-upFree, Plus or Pro; new, renew, upgradeSession JSON, precheck, then payUsually 5–30 minutes
Codex credits (500 per unit)Plus or Pro; not FreeSession JSON, precheck, then payUsually 5–30 minutes
Plus ready-madeNo account neededPay, enter receiving mailbox, account sentReady to use on delivery since 11-09; page shows about 30 minutes to 4 hours
Pro 100 / 200 / 500 ready-madeNo account neededPay, enter receiving mailbox, staff send the accountAbout 30 minutes to 4 hours
Business seatsCompany purchase, minimum two seatsStaff delivery; invoice via support emailAgreed with support

ClaudeMax FAQ (English)

9. Ordering a top-up: the Session precheck, step by step

Every top-up on this site follows the same sequence, and the sequence is designed so you pay only after the system has confirmed that your account can take the product. The credential you submit is the ChatGPT Session or Cookie JSON the order page asks for; it is used for this one top-up and cleared after delivery, your password is never requested, and the platform does not read your conversations. Export it from a desktop browser while signed in to chatgpt.com, not from any other site: the order API rejects credentials from the wrong domain and names the site, which is the most common first-attempt error we see.

Credit packs follow the same steps with one extra field, the number of units from 1 to 20 chosen with the minus and plus buttons on the product card; the total is unit price times units. Credits land on your own account, are used only after your included allowance is exhausted, and are valid for 12 months under OpenAI's rules.

  • Before exporting, confirm your account state against section 8: Free or expired Plus for a Plus top-up; Free, Plus or Pro for any Pro tier; Plus or Pro for credit packs.
  • Turn off Fast and Ultrafast in Codex if you want the pack or allowance to last; both multiply consumption.
  • After delivery, check Settings, My Plan on chatgpt.com (or Usage and Billing in Codex for credits) and keep the order number for renewal.
  • Payment outside China is in USDT on TRC20 or BEP20 at a fixed rate shown on the order page; the checkout and all order emails are in English.
  1. Pick the product from the decision table and open it on the pricing page
  2. Sign in to chatgpt.com in a desktop browser and export the Session or Cookie JSON the order page asks for
  3. Paste it on the order page; the system prechecks account state and tier eligibility
  4. Pay only after the precheck passes (USDT TRC20 or BEP20 outside China)
  5. Delivery usually within 5–30 minutes; verify in Settings, My Plan or Usage and Billing
  6. Record the order date for the 30-day renewal reminder
The top-up flow on this site: nothing is charged until the precheck passes, and the credential is cleared after delivery.

OpenAI Help Center: Using Credits for Flexible Usage in ChatGPT (Personal plans) · ClaudeMax FAQ (English)

10. Renewal: 30-day cycles, same-tier pricing, and an offline reminder

Top-ups on this site run in 30-day cycles. Renewing a top-up means placing a new top-up order of the same tier; the site emails reminders 7, 3 and 1 days before expiry. Ready-made accounts renew at the same-tier top-up price: outside China you send the order ID via Telegram, 30 days are added and the account is unchanged. Credit packs have no expiry on this site's side because they are one-off purchases; on OpenAI's side the credits are valid for 12 months from purchase and do not roll over.

OpenAI Help Center: Using Credits for Flexible Usage in ChatGPT (Personal plans) · ClaudeMax FAQ (English)

If you prefer not to rely on email, the snippet below computes the expiry and a re-order date from the order date and prints a crontab line that writes a reminder to a log file on the reminder morning. It runs offline, touches no account and sends nothing anywhere.

renew_reminder.sh — offline; works on macOS (BSD date) and GNU/Linux
#!/usr/bin/env bash
# renew_reminder.sh - offline reminder for a 30-day top-up. No network, no account access.
# Usage: ./renew_reminder.sh 2026-11-10 "ChatGPT Pro 200 top-up"
start="$1"; label="${2:-subscription}"
if date -j >/dev/null 2>&1; then                                 # macOS / BSD date
  expiry=$(date -j -v+30d -f "%Y-%m-%d" "$start" +%Y-%m-%d)
  remind=$(date -j -v+27d -f "%Y-%m-%d" "$start" +%Y-%m-%d)
else                                                              # GNU date
  expiry=$(date -d "$start +30 days" +%Y-%m-%d)
  remind=$(date -d "$start +27 days" +%Y-%m-%d)
fi
echo "$label: started $start, expires $expiry, re-order on or before $remind"
month=$(echo "$remind" | cut -d- -f2 | sed 's/^0//')
day=$(echo "$remind" | cut -d- -f3 | sed 's/^0//')
echo "crontab line (09:00 on the reminder day):"
echo "0 9 $day $month * echo 'Renew $label before $expiry' >> \$HOME/renewals.log"

One renewal trap is specific to Pro 200. Because OpenAI now treats all Pro 200 subscriptions alike, there is no longer an old-allowance account worth preserving: a lapsed Pro 200 that you resubscribe gets the same Work and Codex allowance as everyone else, and an active one does not need to be canceled and re-bought to receive the doubled allowance. Do not pay anyone a premium for a grandfathered Pro 200.

OpenAI Help Center: About ChatGPT Pro tiers

11. Risks, refund wording, and a pre-order self-check

The vendor-side risks are the ones OpenAI states. Sharing credentials, reselling access and automated extraction are prohibited under the Terms of Use and can lead to restriction; a usage limit is not a restriction and support cannot reset limits. Credits are non-refundable on OpenAI's side except for verified billing errors, confirmed unauthorized activity or service failure. Pro has no annual billing, so there is no annual discount to miss.

OpenAI Help Center: About ChatGPT Pro tiers · OpenAI Help Center: Using Credits for Flexible Usage in ChatGPT (Personal plans)

This site's refund terms for new orders, as published in the FAQ: ready-made accounts are not refundable once delivered; losses caused by account bans or restrictions are not covered by refunds or compensation; when the account works normally there is no seven-day no-questions or mid-month cancellation refund. For ready-made accounts the operator's phrase is that the subscription is guaranteed for the term, the account against bans is not. If anything goes wrong, contact support promptly through the chat in the bottom-right corner. ClaudeMax is not affiliated with OpenAI or Anthropic.

ClaudeMax FAQ (English)

Before you order, fill in the checklist below honestly. It encodes the eligibility rules from section 8 and the Ultrafast rule from OpenAI, and its last line is the only question that matters: does your current plan appear in the list the product accepts.

preorder_check.json — fill in the account block, then compare it with the rules block
{
  "checked_at": "2026-11-10",
  "account": {
    "can_sign_in_on_desktop": true,
    "current_plan": "plus",
    "plan_expires": "2026-11-24",
    "billing_platform": "web",
    "two_factor_enabled": true,
    "codex_limit_hits_last_30_days": 9
  },
  "want": "pro-100",
  "rules": {
    "plus_topup_accepts": ["free", "expired-plus"],
    "pro_topup_accepts": ["free", "plus", "pro"],
    "credit_packs_accept": ["plus", "pro"],
    "ultrafast_only_on": "pro-500",
    "credits_unlock_ultrafast": false
  },
  "ready_to_order": "account.current_plan is in rules.pro_topup_accepts"
}

What we did not verify: we did not measure any Pro tier's Work or Codex allowance before or after 2026-10-29, because OpenAI publishes no numbers and we hold no instrumented accounts; the estimator's task sizes are placeholders, not observed token counts; we did not test Ultrafast speed or confirm the up-to-8x figure beyond the vendor's documentation; we did not check OpenAI's own credit pack prices, which vary by account and region; and the gHacks quote is a press report, not a first-party page. Help center pages were read through a text proxy on 2026-11-10 because they block direct fetches. Prices in this article are live values from this site's catalogue and are not compared with OpenAI's list prices. ClaudeMax is not affiliated with OpenAI.