The same thing is happening to me, and now I’m stuck with Terra. It’s not bad, but I have to manually verify every step, which is exhausting.
Thank you to everyone who has reported experiencing the same problem.
Note: I am not an OpenAI employee, just a user of the forum.
OpenAI is working to investigate and resolve the issue, but they need additional information from those affected. Please include the following details in your reply:
Additionally, could you share the Feedback ID or Request ID with me? That would really help with the investigation.
Avinash
019fa9cd-775c-73f3-a23b-982d922481ae
If there isn’t enough capacity to support the service reliably, perhaps it would be worth reconsidering whether it should continue being sold as if it were fully available.
I have a total allowance of 100 million tokens, yet my task has been running for two days and has been interrupted repeatedly by the same capacity error. At this point, I’m genuinely curious how users are expected to complete any serious work under these conditions.
I paid for a membership to use the service, not to spend my time repeatedly looking at error messages. A paid subscription should provide access to a functioning product, not simply a more consistent view of the same failure screen.
The issue has now persisted for two full days. One would assume that a capacity problem affecting paid users this consistently would receive rather urgent attention, but apparently that expectation may have been overly optimistic.
Could you please explain when this issue will actually be resolved, rather than simply asking users to keep trying again?
I’m seeing similar reports from other users, so it may be a temporary capacity issue rather than something specific to your setup. Since you’ve already submitted feedback and contacted support, you’ve covered the right steps. Hopefully OpenAI can confirm whether there’s still an ongoing backend capacity issue with GPT-5.6 Sol. In the meantime, trying another model is probably the best workaround until the service stabilizes.
This issue is currently under investigation, and I'll share an update here if there are any developments.
In the meantime, what really helps the investigation is having the relevant user, feedback, session, or request IDs when trying to reproduce the issue. To keep that information private, please report the issue by emailing support@openai.com, then share the case number here so I can follow up.
Avinash
4 Likes
I am now extremely happy to announce that the Thread Alive Keeper powered by 5.6 Luna is completely unusable which makes 5.6 sol officially useless in anyway I can possibly think of. To keep my job, I just subscribed a Claude Max plan, solely because I have absolutely no way possible to access 5.6 sol whatsoever.
I feel exhausted and frustrated. In a thread (Feedback ID: 019fa50e-3b02-72b3-b560-9056bcf7b3e9) Luna said “continue” a whopping 80 times in ONE single thread and sol didn’t even say a single word. The “Continue Pre Minute” can go up to 8 CPM.
That means Luna was sending “continue” roughly once every 7 seconds at the peak. But none of those retries were progress. They were just new turns being created while Sol remained silent. Now I sit here, watching it keep on going “Selected model is at capacity. Please try a different model.” Then I laughed, I never thought it could be so hilarious.
@OpenAI_Support Thank you so much for this amazing comedy.
I’m experiencing the same issue, and honestly, this has become extremely frustrating.
I have two separate Pro accounts, and GPT-5.6 Sol has been almost unusable on both of them for the past two days. It repeatedly returns:
“Selected model is at capacity. Please try a different model.”
Both accounts still have more than 41% of their Codex usage remaining. Terra, Luna, and GPT-5.5 are all working normally. Sol is the only model that keeps failing, and it happens throughout the day rather than during a single short period of high demand.
A friend in the same region tested Sol with a Plus account at around the same time and was able to use it normally. Since both of my Pro accounts are affected while the Plus account works, I’m concerned that this may be related to certain Pro accounts or a specific account routing group.
I understand that temporary capacity errors can happen occasionally, but this is no longer an occasional issue. It has continued for two days and is now interrupting almost every task I try to run with Sol.
This has caused me to lose a significant amount of time. Tasks are repeatedly stopped in the middle, I have to switch models or restart the work, and I can no longer rely on Sol for anything important. It is especially frustrating because I am paying for two Pro subscriptions, both accounts still have plenty of usage remaining, and yet the model is effectively unavailable to me while other users appear to be using it normally.
Could someone from OpenAI please investigate this as an account-side routing or capacity allocation issue rather than treating it as a normal temporary overload? I can provide exact timestamps, screenshots, Feedback IDs, and Request IDs privately if needed.
I would really appreciate a clear explanation or an actual resolution, because at the moment this is seriously affecting my work.
Yeah, it does seem only affect Pro users, my friends who has plus doesn’t seem to have this kind of problem
New Discovery: I tried to use CLI Proxy API (something Tibo mentioned on twitter) and then connected it to my Codex using CC-Switch, and ALL models are useable in any reasoning effort all of a subbden! It is a life saver, I can now finally use Sol Max and Ultra to get my things done.
Same Issue here. My friend with the Plus account never ran into this issue with my Pro account runs into this issue too many times.
Nope. I have both a Plus and a Pro subscription: one for work and one for everyday use. Both are affected.
@OpenAI_Support we demand answers!!!
Can we get some update before GTA6? @OpenAI_Support
I was able to use every model again without problems. I suppose they fixed it. Also, using it directly from the terminal worked from almos the beginning.
Hey team,
This issue has been fixed on our end. Could you confirm whether you're still experiencing it?
If the issue is still persisting, please report it to support@openai.com.
Avinash
Hi Avinash, thanks for the update.
I will verify whether GPT-5.6 Sol is consistently usable again across multiple projects before confirming that the issue is resolved.
Even if service has now been restored, this incident caused prolonged loss of the paid Sol Max/Ultra capability under my Pro 20X plan. I also consumed substantial additional Luna quota through repeated recovery attempts, including 70 recorded “Continue” inputs, and had to subscribe to Claude Max to keep my work moving.
Could you please confirm what remedy is available? I am requesting a prorated Pro 20X service credit or refund for the affected period, as well as an adjustment for the additional Luna usage caused by this failure. Please associate this request with Support Case #12301157 and handle any account-specific details privately.
Thank you.
