@mark_k@ChatGPT Could you please clarify how to choose between the two model variants, "sunrise" and "flare"? Is there a specific selection process, or is the model chosen automatically?
@kloss_xyz Could you please clarify how to choose between the two model variants, "sunrise" and "flare"? Is there a specific selection process, or is the model chosen automatically?
Could you please explain the specific advantages of upgrading? My primary concern isn't increased usage, but rather avoiding vendor lock-in with other models, which I find unhelpful. Additionally, it was a positive change when the five-hour limit was removed, so its sudden reintroduction is frustrating. Now, we can only complete a small amount of work before being restricted for five hours.
GPT 6 Astra is draining Codex subscriptions.
GPT Astra is the most token efficient model ever built and it still eats a full week in a day.
The only thing keeping Codex usable is OpenAI handing out resets every 48 hours. Without them subscribers get almost nothing.
Fix the limits @OpenAI.
The new 5hr limit makes Codex almost unusable for Plus users
What’s the point when GPT 5.6 Sol on High ( I'm not even using GPT 6 Astra atm) can burn through the 5hr limit before I can even get into a proper coding session?
Let us use the weekly usage freely @thsottiaux
Still broken after a month! 🚨 @OpenAI@OpenAIDevs@thsottiaux@sama asking ChatGPT for a simple text output, it automatically triggers the image generator instead of just responding. Please look into this glitch, it’s completely disrupting the user experience.
One thing that still bothers me is the way Tibo recently implied that Plus users are mostly “casual,” while the real professionals are on Pro.
He literally described Plus users as “relatively casual and new users” when he brought the 5-hour limit back on Plus, while keeping Pro $100 and $200 exempt for now.
I like Tibo a lot and I genuinely think he’s brilliant at what he does, but that framing was unfortunate.
That framing can easily sound like “professional” means either:
→ you code all day
→ you run a company that burns insane amounts of tokens
→ you’re rich enough to treat compute like a rounding error
But there are teachers, researchers, designers, writers, analysts, engineers, architects, lawyers, doctors, scientists, indie developers, artists, translators, consultants, journalists, small business owners, and a thousand other professionals who can absolutely depend on AI every day without needing a $100–$200 plan.
Paying $20/month does not make someone a casual user.
I really hope OpenAI starts valuing Plus users more as the new model variants roll out, with better features, stronger models, and products designed for serious users who simply don’t live inside the highest compute tier.