A few observations after using Grok @bot for a bit. Something about it is special. It has this human wit to it that I don't see anywhere else other than Gemini models (which no one cares about). It is also SUBSTANTIALLY more creative than any other model I have tested. That's it.
@SnazzyLabs at this exact same time, i tried signing into Microsoft with 6 different 2fa methods, and not a single verification code or passkey worked. Appreciate apple while you have it.
going to be completely honest. For most people, code improvement past this point we are at right now is negligible. If you prompt something, the frontier models will get it done, itβs just a measure of quality at this point, which isnβt directly due to a better coding score on a benchmark.
GPT-6.2 Sol & Astra are *finally* here. We've also expanded ultrafast to Pro 200 plans for GPT-6.2 Sol, it's my go-to model for most day to day things.