Gemini 4 is too expensive and caching is opaque, a GPT-6.1 Sol user says
A user who mainly relies on GPT-6.1 Sol says Gemini 4 is currently expensive with unclear caching, treats it as a backup model like Grok, and looks forward to Gemini 4 Flash.
Oh no I ran out. The model is ok. Keep in mind I'm not a Claudescep who uses Anthropic models. So I'm like 3-5 months behind all of you. I'll still be using GPT 6.1 Sol mostly cause it goes further and Tibo resets. This thing is currently expensive. I'll be happy when we have Gemini 4 Flash. You have to beat it with a stick. I don't trust Google's use of caching, so I just assume each reply is not hitting the cache. I'm going to treat this model like I treat Grok. It's for when Tibo reset wave stops and I have to use other models to carry my OAI usage to the end of the week. So long as Tibo resets happen, I barely touch the other models.
This post quotes another post: x.com/presidentlin/status/2111044681761177939