User compares Opus 5.5 with Astra, says Astra usage is hard to sustain
A user says ChatGPT models feel primitive after using Opus 5.5, that Sol 6.1 needs constant correction, and that Astra is good but its usage limits make it hard to use sustainably.
After using Opus 5.5 since launch, ChatGPT models are starting to feel so primitive The difference isn't just intelligence. It's how well Opus understands what you want without needing constant guidance. With Sol 6.1, I keep having to explain, correct, and iterate. Opus 5.5 just seems to get it, and the creativity is on another level. And the crazy part? On the $20 plans, both can burn through a 5-hour limit pretty quickly, but I get so much more actual work done with Opus. Astra is great, but the usage makes it hard to use sustainably. Atpt, Sol feels more like a subagent compared to Opus 5.5.😭😭LOL Anyone else feeling this difference?