Thinking of turning my old gaming PC into a self-hosted GitHub Actions runner. This might benefit me with persistent caching.
1
1
33
I really don't see anything appealing about this design.
Some people might find it useful, but honestly, I don't want to see a wall of threads running at different stages.
I'm already swamped with work and an endless backlog of things to get through. This just adds another layer of overwhelm.
The whole point of delegating work is to start it and forget the threads exist until they're done.
Otherwise, you'll end up spending more time watching threads progress than actually getting other work done.
Every single time someone says T3 Code isn’t “optimal for real work” they show the most useless 2025-era kanban slop that wastes over half the screen real estate for in-progress work.
This is the most poorly optimized UI for real productivity that I think I’ve seen in my life.
2
84
Replying to @gbrancaglione_
Thats true i have felt on many projects one it worked so well i love it and the next day i hate like its working like that
1
11
90,000 tokens just to say GOOD MORNING 😭
Did the math and it gets even crazier:
- 16 seconds of runtime
- ~7,438 tokens/second
- $0.01 reported cost
- 8% of my 5hr Plus limit gone
That's potentially 12–13 good mornings before my entire 5hr quota is wiped LOL!??
and this was GPT-6.2 sol on LOW, with no browsing, coding, or file operations. Either Codex is counting a ridiculous amount of hidden overhead, or something is seriously wrong with the usage accounting.
What's going on chat?
58
i really really don't like the new herdr ui man
bad font, bad icons, layout shifts a lot
nothing beats t3code
64
ChatGPT Astra 6.2 is now showing in config files
It is now showing in some config files, that means it is getting closer, based on similar configs before, a week away maybe?
canceling the plan RN
Did they change 6.1 sol pricing ?
this account was on $20 plan 6.1-sol-low
@thsottiaux @OpenAI
14
Did they change 6.1 sol pricing ?
this account was on $20 plan 6.1-sol-low
@thsottiaux @OpenAI


















