Post

@code_coded Seeing the same lag spikes, especially mid-afternoon ET. Usually clears up if you switch to a different model in the same session (4o works, then back to 5.5). Not a fix, but buys you 20 minutes when you're stuck.
English

@code_coded No, not at all, it has been great for me. In a way it can kinda be slow, but I find other areas to work on, and I’ll sometimes have some concurrent task going on at the same time. What I have found is the accuracy is great compared to before.
English

@code_coded You're not alone, I'm having the same problem. Codex broke my working application. 😢
English

@code_coded I just completed one task this afternoon and cost my 10% weekly lmao
English

@code_coded Yes, super slow and slightly dumber. Hopefully this means 5.6 is coming, lol
English

@code_coded Tried 5.4 on extra high intelligence standard speed, it is slowest and credits consuming model too. Taking too much time on compacting context.
English

@code_coded same here, it feels like it's cutting corners everywhere
English

@code_coded honestly this is why we built maxagents. you shouldn't have to pick one model, just let the agent route to whatever works best
English

@code_coded They’ve been facing latency issues I think, and still under repair
Hopefully fixed soon
English

@code_coded no changes in speed here, but it performed worse over the past few days for some reason. it seems to be back to normal, though.
English

@code_coded So true. Just cancelled claude for codex and it started draining token and less intelligent
English

@code_coded Agreed and you need to have both but unfortunately both now feel so much dumber that 1 week ago.
English

@code_coded 5.5 was trash from day one of release. I deeply hate that model
English


