Pusher

Honestly, been a slow month for me. Surprised it's higher than you guys. I'd expect y'all who maintain the software my businesses relies on to be tokenmaxxing more than me.

Literally no idea how anyone else is using ChatGPT connection on nightly. I've had to disconnect and re-connect 50 times in the past week. This PR looks related: github.com/pingdotgg/t3code/…

Decided to give T3 Code a go and just constantly running into issues (same as I did a few weeks back)... Am I doing something wrong @theo ? Disconnecting and reconnecting works. But this is frustrating!

@theo How do I choose "fast" for codex models in T3 Code?

@theo Do I need T3 Connect if I'm running everything on the same laptop (do I need to run the `npx t3 connect` command on my macbook that is connected above)? What if I want to use the app? Might be helpful to make the instructions a little clearer here?

@thsottiaux @poteto "We've reset your usage to 100% of something we can't supply you with." Scrap the 28 day gameplan my guy. There should be 2 items on your todo list. - How do we compete with Opus 5.5 - How do we get more compute.

Is this the wrong asset on the blog or does the dashboard back button and breadcrumb actually have the weird alignment and spacing?

Surely with AI we can find a solution to dealing with this daily @poteto

That's new. @thsottiaux

👀Reset when? Put my stuff on Ultra+fast mode as I was expecting a reset today 😬

GPT5.5 really does love that specific style infobox lol.

Forgot the screenshot...

That must be satisfying.