Search
Filter by tool, plan, model, or the kind of report. A search that matches nothing is logged.
Searching…
How long the plans actually last.
Filter by tool, plan, model, or the kind of report. A search that matches nothing is logged.
Searching…
56 reports.
“$250 account weekly budget suddenly gone in an hour on a relatively low workload”
Weekly budget gone in an hour on a relatively low workload on Oct 14; next day back to normal under heavier load.
Hit the weekly capCodex (plan not stated)(ChatGPT ($250 account, tier not stated))philemonhollandgithub.com
“I ran out of quota on both bars this month less than a week ago”
Ran out of both usage bars, then used slower courtesy Auto until it ended on Oct 11.
Hit the monthly capCursor Student(Cursor (student plan, $20))Autogseforum.cursor.com
“Previously, similar Sol Ultra sessions typically lasted around 1–2 hours before I reached the usage limit.”
5-hour allowance exhausted in ~2-3 minutes; similar sessions previously lasted 1-2 hours.
Hit the 5-hour capChatGPT PlusSol Ultrakautuksahnigithub.com
“my weekly limits are going way faster than last week”
Weekly and session limits draining much faster than the prior week while using a single conversation.
Faster drainClaude (plan not stated)gysosingithub.com
“a task involving two images reduced my remaining allowance from 100% to 16%”
Astra Max task with two images: 100%->16% of 5-hour; one question 100%->80%; five Sol questions left 67%.
5-hour window costChatGPT Plus(ChatGPT Plus (Work))GPT-6 Astra Max / GPT-6.1 Solgus6443-beepgithub.com
“My weekly limit now runs out in about two days, while the same work used to last about a week”
Weekly limit now runs out in about two days; same work used to last about a week.
Hit the weekly capClaude Max 20x(Claude Max 20x (EUR 200/mo))BoozyWoozy88github.com
“astra high burned the 5h usage limit in 10 minutes”
Astra high burned the 5-hour limit in 10 minutes before editing any code.
Hit the 5-hour capCodex (plan not stated)GPT-6 Astra (High)erw-1github.com
“the weekly limit now runs out in about half the time”
Weeks starting Sep 3 and Sep 10 hit the weekly limit after 6d16h and 6d18h; now runs out in about half the time with less work.
Hit the weekly capClaude (plan not stated)(Claude subscription (tier not stated))Opus 5.5Vlastelynagithub.com
“The more recently charged accounts? $125”
Early-September Ultra accounts got $500 of third-party model usage; more recently charged accounts got $125.
Pool size noteCursor UltraAlien_Gateforum.cursor.com
“Same pattern here, on a Team seat rather than Max.”
Same faster weekly drain pattern on a Team seat with unchanged workload.
Faster drainClaude Team(Claude Team (Max 5x tier))Opus 5.5mattoneygithub.com
“the entire weekly Codex allowance is being consumed within roughly 5 hours of intensive but normal development work”
Entire weekly Codex allowance consumed within roughly 5 hours of intensive but normal development work.
Hit the weekly capChatGPT Pro(ChatGPT Pro 5x ($100))GPT-6 Astra1337Mizugithub.com
“weekly limit at 93%”
Weekly limit at 93% about 36 hours after the Sep 26 reset.
Near the weekly capClaude Max 20xOpus 5.5 (high/xhigh, many subagents)yurkomikgithub.com
“I now hit 38% of session limit during a single planning task”
38% of the 5-hour session limit used during a single planning task.
Near the 5-hour capClaude Max 5xsavagergithub.com
“my 5 hour usages limit on plus plan gone in 2 minute of astra high model thinking”
5-hour limit gone in about 2 minutes of Astra high thinking.
Hit the 5-hour capChatGPT PlusGPT-6 Astra (High)l33tx0github.com
“"Weekly - all models" is at 97%”
Weekly (all models) at 97% within about one day after upgrading to Max 20x.
Near the weekly capClaude Max 20xOpusantiloveBabagithub.com
“the previous week took 9,352 responses to reach 100%”
Previous week took 9,352 responses to hit 100% weekly; after Sep 25 reset ~30 responses per 1% vs ~93 before (~3.6x faster).
Hit the weekly capClaude (plan not stated)(Claude (plan not stated; has usage credits))konradkozakiewicz-dotgithub.com
“$24.55 of actual usage is showing as 26% of my Other Models usage”
$24.55 of actual usage shows as 26% of Other Models right after upgrading.
Model pool rateCursor Ultranhughforum.cursor.com
“Other Models is at 100% after about $127 of Opus”
Other Models at 100% after about $127 of Opus at published rates.
Hit the model poolCursor UltraOpusJin_Zhangforum.cursor.com
“After today’s GPT-6 Sol and Luna release, credit usage feels dramatically better.”
After the GPT-6 Sol and Luna release (Sep 23), credit usage feels dramatically better than Astra Light.
Usage improvedChatGPT BusinessGPT-6 Sol / GPT-6 Lunapirolidgithub.com
“I only had 11% left, after just two prompts”
Two prompts (two web pages to PowerPoint) left only 11% of the 5-hour limit.
5-hour window costChatGPT PlusGPT-6 Astra (Medium)raidologithub.com
“The public-rate sum is about ~$260”
Other Models meter at 100% while usage at published API rates reconstructs to only ~$260 (~65% of a $400 pool).
Hit the model poolCursor UltraGemini + SonnetMatomforum.cursor.com
“After ~2 hours in a single Agent chat”
Hit the Other Models usage limit after ~2 hours in a single agent chat; Cursor switched models.
Hit the model poolCursor (plan not stated)Third-party modelsSAK_CBforum.cursor.com
“the All Models weekly usage indicator jumped to 100%”
Weekly "All Models" indicator jumped to 100% immediately after upgrading from Max 5x to Max 20x.
Reported meter issueClaude Max 20xcapitaletechgithub.com
“about 10 % in the first ~10 hours on 20x, versus about 60 % in ~41 hours on 5x the weekend before”
~10% of weekly used in the first ~10 hours on Max 20x vs ~60% in ~41 hours on Max 5x the weekend before.
Faster drainClaude Max 20xerplandgithub.com
“it’s the 14th and I’m already at 99% usage”
Plan no longer lasts half the month; on the 14th already at 99% usage on Composer/Auto.
Near the monthly capCursor ProComposer / AutoCocina_facil_con_juaforum.cursor.com
“every 100 million tokens would increase by 10%”
Using Auto, every ~100 million tokens raised usage by about 10% (Aug 19-Sep 19 cycle).
Monthly rateCursor ProAutojie_yuforum.cursor.com
“2 prompts and the whole 5H session is gone”
2 prompts used the whole 5-hour session on Astra; similar prompts on Sol used ~10%.
Hit the 5-hour capCodex (plan not stated)GPT-6 Astra vs SolEnigmaCEOgithub.com
“I burned through 1,500 credits in just six hours after the 5.6 update”
Burned 1,500 credits in six hours after the 5.6 update; previously used ~200 credits a month.
Hit the monthly capGitHub Copilot(Copilot (1,500-credit plan))Donny1717github.com
“Now just in one day I've gone through all my available usage.”
Went through all available usage in one day, and maxed out paid extra usage too.
Hit the monthly capGitHub Copilot(Copilot (annual subscription))Sonnet 5github.com
“I used my week limit + 2 resets within a couple of days”
Used the week limit plus 2 resets within a couple of days; 11% left for the next 5 days.
Hit the weekly capChatGPT Pro(ChatGPT Pro 5x)GPT-6 Astra (High/Medium)swisswebxpertsgithub.com
“it consumed my quota that last for 30 days in just 1 week”
Quota that normally lasts 30 days used in about a week; staff says Grok 4.6 High was picked manually and heavy use came earlier.
Hit the monthly capCursor ProAuto / Grok 4.6 HighNanya_Airconnforum.cursor.com
“Time until quota was exhausted: approximately 4 minutes / one prompt”
5-hour quota exhausted in ~4 minutes / one prompt; with GPT-5.6 Sol the same window gave 30 min to 1h+.
Hit the 5-hour capChatGPT PlusGPT-6 Astra (Medium) via OpenCodeDavidtrassugithub.com
“5-hour quota now exhausts in ~86 minutes”
5-hour quota now exhausts in ~86 minutes despite lower token activity (around Sep 9).
Hit the 5-hour capChatGPT Plus(Codex CLI)TomyFovosgithub.com
“Not consume 50% of the 200 USD account in a few turns of doing nothing.”
Weekly went from ~60% to 12% remaining in under 5 minutes of light work (Sep 9 incident).
Reported meter issueChatGPT Pro(ChatGPT Pro ($200))GPT-6 Astra (XHigh)redknightloisgithub.com
“I went from 100% to 0% after 3-4 messages, around 15min work!!!”
Went from 100% to 0% after 3-4 messages (~15 min) during the Sep 9 meter incident; briefly restored then dropped again.
Reported meter issueChatGPT Plusdjoolegithub.com
“After roughly 2.5 minutes, about 50% of my 5-hour Codex usage window had been consumed.”
Codex Security scan of a small repo used ~50% of the 5-hour window in ~2.5 minutes.
5-hour window costChatGPT PlusGPT-6 Astra (High)iamspidogithub.com
“a single very short GPT-6 Astra interaction consumed approximately 6% of my 5-hour allowance”
One-sentence Astra prompt used ~6% of the 5-hour allowance; ~30 minutes of GPT-5.6 used ~25%.
Cost per messageChatGPT PlusGPT-6 Astra vs GPT-5.6NUS-Timgithub.com
“fresh allowance: 0% -> 83% used in only 11h23m”
First weekly allowance 98% consumed in ~30-40 hours; fresh allowance 0%->83% in 11h23m.
Hit the weekly capCodex (plan not stated)x997hubgithub.com
“GPT-6 Astra consumed essentially the entire 5-hour Codex allowance in two very short, ordinary coding turns.”
Two short turns: 100%->42% (<5 min) then 42%->0% (1-2 min), hit usage limit before finishing.
Hit the 5-hour capChatGPT PlusGPT-6 Astra (Medium)GustavoMartins123github.com
“My entire 5-hour allowance was exhausted after only 3 normal prompts”
Entire 5-hour allowance exhausted after only 3 normal prompts.
Hit the 5-hour capChatGPT Plusjgcrostagithub.com
“The first task used an entire 5 hour budget and 16% of the weekly”
First task used the entire 5-hour budget and 16% of weekly; on Plus a 5-hour window is ~16% of the weekly total.
Hit the 5-hour capChatGPT PlusFromArielgithub.com
“took roughly 45 minutes and consumed about 48% of the 5-hour window”
One ~45-minute Codex task consumed ~48% of the 5-hour window.
5-hour window costChatGPT PlusLucifor1976github.com
“I hit the weekly limit at roughly the same point in the cycle regardless of tier”
Hits the weekly limit at roughly the same point in the cycle on Max 5x and Max 20x; on 20x never hits the 5-hour cap.
Hit the weekly capClaude Max 5x vs 20x(Claude Max 5x / Max 20x)dlaszlogithub.com
“Within only a few hours, roughly 50% of my remaining Codex usage quota suddenly disappeared”
~50% of remaining Codex quota disappeared within a few hours; reset window moved forward a week.
Faster drainChatGPT Proamiras2030github.com
“this month: 3 prompts, 7000 credits consumed!”
This month: 3 prompts consumed 7,000 credits.
Cost per messageGitHub Copilot(Copilot (plan not stated))github.com
“only running about 2 sessions parallel”
Close to the 5-hour session limit while running only about 2 parallel sessions with nothing heavy.
Near the 5-hour capClaude Max 20xamidan99github.com
“34 hours into the first full Max 20x week”
First-ever weekly limit hit 34 hours into the first full Max 20x week (after upgrading from Max 5x).
Hit the weekly capClaude Max 20xcheonnamsugithub.com
“I burned through my entire Ultra allowance in just 4 days”
Burned through the entire Ultra allowance in 4 days while setting up an SDLC workflow.
Hit the monthly capCursor UltraStanislav_Chistyakovforum.cursor.com
“0% → 11% → 59% → 100% in about 18 minutes”
During an idle watch, 5-hour session went 0%->100% in ~18 minutes and weekly 57%->72% without a message sent.
Reported meter issueClaude Max 20xakshayramdassgithub.com
“Weekly usage quota and Fable quota both jumped from 0% to 50% / 100% respectively while my account was completely idle”
Weekly jumped 0%->50% (all models) and Fable to 100% while the account was idle; ~40 real messages later moved it only 50%->54%. (Anthropic later told another user in this thread that idle drains were likely stolen sessions.)
Reported meter issueClaude Max 20xOpusmoepowersgithub.com
“Session (5hr) 6% / Weekly (7d) 2%”
After ~1 hour of active work from zero: Session 6% / Weekly 2%.
Weekly ratioClaude Max 20xOpus (high effort)Regsormgithub.com
“the weekly allowance is now exhausted after roughly 3x less work than before”
5-hour:weekly consumption ratio moved from ~10:1 to ~3:1, so the weekly allowance runs out after ~3x less work.
Weekly ratioClaude Max 20xOpus (high effort)macblackstuffgithub.com
“on Max 5x, ~4-5 Opus messages consumed 1% of the weekly Opus limit and my allowance lasted 6-7 days of normal work”
After a mid-cycle upgrade to Max 20x, ~2 Opus messages use 1% of the weekly limit; on Max 5x it was ~4-5 messages per 1% and the allowance lasted 6-7 days.
Cost per messageClaude Max 20xOpusRemy-authoritygithub.com
“the "20x" plan is misleading and only give you twice the weekly limits of the "5x" plan”
Says the "20x" plan empirically gives only about twice the weekly limits of "5x".
Weekly value noteClaude Max 5x vs 20x(Claude Max 5x vs Max 20x)dannywnews.ycombinator.com
“sat and thought for 10 minutes before maxing out our company's per-user weekly limit”
Coworker's Sol run thought for ~10 minutes and maxed out the company's per-user weekly limit.
Hit the weekly capChatGPT company plan(ChatGPT company plan (tier not stated))Soldoawoonews.ycombinator.com
“the Pro 20x $200 plan gives you about $2200 API-equivalent weekly usage if you're only using GPT 5.6 Sol”
Estimates the $200 Pro plan gives about $2,200 API-equivalent weekly usage on GPT-5.6 Sol.
Weekly value noteChatGPT Pro(ChatGPT Pro 20x ($200))GPT-5.6 SolTiberiumnews.ycombinator.com