Is a lot of money. The majority of people here aren't willing to spend $200/mo for coding unless their little projects provide a comparable value back to them.
For context, I'm paying under $30/year and get GLM-5.2. An extra $2300/year isn't going to get me much better outcomes.
The point is that the less capable models are also borderline AGI. You're paying 10-100x more to get a few percentage points improvement in performance.
Put another way, what I get for my under $3/mo is better than what you were getting 3-5 months ago paying $200/mo. So you're paying a lot just to be ahead by a few paltry months.
Is a lot of money. The majority of people here aren't willing to spend $200/mo for coding unless their little projects provide a comparable value back to them.
For context, I'm paying under $30/year and get GLM-5.2. An extra $2300/year isn't going to get me much better outcomes.
it seems cheap for what is borderline AGI
The point is that the less capable models are also borderline AGI. You're paying 10-100x more to get a few percentage points improvement in performance.
Put another way, what I get for my under $3/mo is better than what you were getting 3-5 months ago paying $200/mo. So you're paying a lot just to be ahead by a few paltry months.