It’s impossible to get any meaningful work done. GPT Sol is an improvement over 5.5 but it now blocks work for hours sometimes days before it finally returns results which delays that tight prompt-fix cycle.
On top of that on the $200/month I get at most 2~3 days of use with Sol Medium. I’ve burned through all the banked resets.
Just reading the threads about this online, I can see that discontent is bubbling up and I think it’s urgent that OpenAI make a statement or a plan at least.
The new usage model has seriously reduced – we don’t even use the 5.6 stuff but on a few items and the week burned through in 3 days – this never happened before - there is a serious usage tracking change or issue
Since the release of GPT-5.6, the five-hour usage window on the $200 plan was exhausted almost immediately, followed by nearly all of my weekly token allowance.
At the same time, the model takes noticeably longer to respond, while the output quality is not better than 5.5.
This is not acceptable for a plan at this price point. The practical difference between the $200 subscription and lower-priced plans is barely noticeable.
Support has not resolved the issue either. My tickets receive only generic responses, with no clear explanation or concrete action.
In its current form, the subscription does not justify its cost, so I am considering alternatives with transparent limits and responsive support.