FULL STORY

GPT-5.6 Sol Quota Issues and OpenAI's Fixes

After GPT-5.6 Sol's launch, rapid quota depletion sparked controversy. OpenAI repeatedly reset limits and optimized the model, ultimately reducing resource consumption by 18%.

2026-07-10 ~ 2026-07-30 · 10 episodes · 91 posts

Episode 1 · GPT-5.6 Sol High-Speed Quota Depletion Sparks User Complaints (2026-07-10, 5 posts)

Recently, the quota depletion rate of the GPT-5.6 Sol model (including High and Ultra modes) has sparked significant user complaints. Multiple users report that previously generous subscription quotas are now being exhausted in a very short amount of time, severely disrupting daily development workflows.

Abnormal Quota Depletion and Degraded Experience

Several users pointed out that the new model consumes quotas at an abnormally fast rate. User @kalmankantaja stated that while using Codex for code refactoring, a 5-hour limit was exhausted in just 15 minutes, with the system warning that less than 10% of the daily quota remained. User @vista8 mentioned that many have been affected by the 5-hour reset rhythm of Codex and Claude Code, noting that GPT-5.6 Sol depletes quotas in just dozens of minutes under High and Ultra modes. User @jdjohnson also reported hitting the 5-hour limit very quickly under their current model setup.

Limited Parallelism and Cost Concerns

The most direct impact of this change is on multi-agent parallel tasks. User @petrusenkomax pointed out that the $200/month Codex plan used to have plenty of headroom, easily handling 10+ parallel agents without hitting the limit. However, under GPT-5.6 Sol, running just 1 or 2 agents often triggers the hard limit. Feedback relayed by @RileyRalmuto corroborated this, noting that running multiple loops simultaneously rarely touched the quota before, making the $200 subscription feel "very worth it"—a stark contrast to the current situation. Furthermore, regarding @sama's claim that this model addresses enterprise cost concerns, @jdjohnson expressed disagreement, stating that the actual experience runs completely counter to those claims.

Episode 2 · OpenAI Repeatedly Resets Codex and ChatGPT Work Limits After GPT-5.6 Sol Launch (2026-07-10, 12 posts)

With the release of GPT-5.6 Sol, OpenAI reset the usage limits for ChatGPT Work and Codex multiple times within 24 hours to encourage users to try more challenging tasks. However, the rollout faced issues; the team acknowledged problems such as high-compute modes consuming limits too quickly, and the frequent resets sparked user discussions regarding quota mechanics.

Official Response and Adjustments

The OpenAI team reviewed the initial feedback and admitted a few missteps in the new update: the highest compute mode was too easy to trigger and consumed too much quota. To address this, they reset the limits and promised another adjustment later that day. Additionally, to clear up user confusion that Codex might be phased out, OpenAI clarified that focusing on ChatGPT Work did not mean abandoning Codex, emphasizing that Codex is here to stay.

User Feedback and Controversies

The frequent limit resets drew mixed reactions from users. @0xKaiser noted their quota was reset three times consecutively—jumping from 20% back to 100%—speculating that this was either due to extra consumption from the new update or dynamic background policy adjustments. Meanwhile, @FHSS97 discussed strategies for account upgrades, pointing out that utilizing a reset to restore usage to 100% before upgrading to a Pro plan offers better value. @alexcovoeth viewed the weekly reset mechanism with expiration dates as a practical improvement that solves the pain points of quota management.

Episode 3 · OpenAI Temporarily Lifts Codex 5-Hour Limits and Boosts Quotas (2026-07-13, 24 posts)

Over the past 48 hours, OpenAI has made several consecutive adjustments to quotas and product strategies for Codex and ChatGPT Work. Most notably, the company temporarily removed the 5-hour usage limits for Plus, Business, and Pro plans, with users noticing the disappearance of the "5h" UI prompt and a weekly quota reset to 100%. These changes directly improve the continuous availability of coding agents and have sparked significant community discussion.

Key Details and Optimizations

According to widely shared updates, Codex has reached 6 million active users. The OpenAI team optimized inference costs and passed the savings back to subscriptions, increasing GPT-5.6 Sol availability by approximately 10%. Additionally, the team reverted a previous change that had temporarily raised the context limit from 272k to 372k.

User Reactions and Discussions

Multiple users view this relaxation as a major signal of improved product practicality. User @Independent-Spirit36 noted that removing the limit eliminates 5-hour cooldowns, allowing for faster iteration and making the "banking-style" reset more practical. Users like @ExodusGreen took the opportunity to call out competitor Anthropic. However, all available materials emphasize that lifting the 5-hour limit is a "temporary" measure, and whether it will become permanent remains unknown.

4 more related posts →

Episode 4 · Codex Developer Update: GPT-5.6 and Parallel Tasks (2026-07-14, 28 posts)

OpenAI used July 15 to present a new wave of Codex updates aimed at developers. Officially, the company said Codex now has 7M+ weekly active users and has shipped 150+ updates over the past two months, underscoring both rapid iteration and Codex’s move toward becoming a more complete entry point for real development workflows.

Key updates

Across OpenAI’s official posts and a demo by developer relations lead Romain Huet, the headline features were access to GPT-5.6 and Ultra, better parallel task management, a major update to Computer Use, inline editing directly in code and docs, plus PR review, Sites, and mobile usage. @OpenAIDevs also mentioned /goal, but did not provide further detail in the post. Huet’s demo video ended with a brief teaser of an unreleased feature.

Growth and quota information

Beyond the official product posts, multiple reposts and commentary posts said Codex and ChatGPT Work together had reached 8 million monthly active users, and that OpenAI had reset usage quotas again. However, the exact limit policy was described inconsistently. Some reposts said there was still no 5-hour rate limit, while Emmanuel Vivier said GPT-5.6 Sol offered Plus, Pro, and Business users “unlimited use on a rolling 5-hour window,” with weekly and monthly caps still in place.

Community observations

Discussion around the update focused on two themes. First, posters such as @rwang07 said Codex usage clearly picked up after GPT-5.6 launched. Second, content reposted by @steipete said the compaction issue had improved noticeably, with Codex 5.6 Sol less likely during long sessions to lose context, forget the goal, or drift off task. Those observations help explain why this update drew attention beyond a normal feature drop.

8 more related posts →

Episode 5 · OpenAI Resets Usage Quotas for Codex and ChatGPT Work (2026-07-16, 2 posts)

OpenAI has reset the usage quotas for Codex and ChatGPT Work users back to 100% following system stability fixes. Additionally, the company noted that ChatGPT Work has already reached 9 million active users.

Episode 6 · OpenAI Frequently Resets Limits, Upping Codex Daily Cap to 10 Million (2026-07-20, 4 posts)

OpenAI has been unusually active in resetting usage limits for ChatGPT and Codex, officially raising the daily cap for paid users to 10 million. The community speculates this massive capacity increase for coding scenarios might be due to freed-up compute resources.

Episode 7 · OpenAI Resets User Limits Following Global Outage (2026-07-25, 4 posts)

Following a near-global outage that caused high error rates in ChatGPT and Codex, OpenAI deployed mitigations and restored services. The company has also reset usage limits for all affected Codex and ChatGPT Work users.

Episode 8 · OpenAI May Hit Compute Limits as ChatGPT Work and Codex Surge (2026-07-26, 2 posts)

ChatGPT Work has officially surpassed Codex in active users, and their rapid growth to 10 million users in four months suggests OpenAI may be hitting compute capacity limits.

Episode 9 · OpenAI Resets Codex Limits and Investigates GPT-5.6 Sol Usage (2026-07-28, 7 posts)

OpenAI recently reset usage limits for ChatGPT Work and Codex paid users and is investigating the issue of GPT-5.6 Sol consuming quotas too quickly. The official clarification states that this is not a reduction in subscription quotas, but rather an experience change brought by system fixes, claiming that the fixed model can last 18% longer.

Confirmed

  • OpenAI has reset usage limits for paid users of Codex and ChatGPT Work.
  • The official team is investigating reports of GPT-5.6 Sol consuming quotas too quickly, explicitly stating that this is not a subscription quota reduction, but an experience change following a fix.
  • OpenAI stated that relevant fixes could improve the token efficiency of GPT-5.6 by about 18%.
  • The previously temporarily suspended 5-hour rate limit has been announced to be restored, but it will remain temporarily suspended while the GPT-5.6 Sol issue is being investigated.

Why it matters

  • This event directly affects the actual available quotas and workflow continuity for ChatGPT Work and Codex paid users. OpenAI's timely reset of limits and suspension of the 5-hour restriction aims to protect the user experience while investigating abnormal model consumption, preventing users from being unable to use the service due to abnormal usage. Meanwhile, the expected improvement in token efficiency brings a positive signal for developers.

Episode 10 · OpenAI Optimizes Codex: GPT-5.6 Sol Usage Limit Extended by 18% (2026-07-30, 3 posts)

OpenAI has optimized Codex by improving operational efficiency, reducing power consumption, and extending the usage limit for the GPT-5.6 Sol model by approximately 18% for Codex and ChatGPT Work users.