I really don't like having to open Work sessions for one-off questions, only because they're limiting what models they put in the chat. The older models are just too dumb for some things.

6.1 is Astra minor. Way more capable but also way heavier+slower. Its really 2 different models, they just shipped it as sol to recover from the gpt6 disaster lunch, where they tried to pass terra 6(or a cheaper model) as sol but it was worse than expected

I’ve hit model not available limits for GPT 6.1 Sol multiple times over the last week. It’s never for very long, and I can switch to Astra, but it seems like OpenAI is struggling with capacity. This has happened with my personal $200 Pro connection and my Codex enterprise connection.

Given the extreme short time between 6 Sol and 6.1 Sol, I suspect they don’t actually have much in common and 6.1 is a heavier model rebranded as Sol in a panic response to poor agentic capabilities of 6.