AI Agent Reviews Code Commits And Runs Multiple Tools In Five Minutes

@LuciferMornens What?

@MatthewBerman Also brutal that they cut off mid-task. Codex allows some extra free over-use to make sure you are not stuck mid-session.

To be fair, I was running 3 tasks in parallel. But still…

@MatthewBerman I fully expect this rerelease to be a shit show.

@doug_fowler @odyth @MatthewBerman weird. i m using fable 5 for a while now with a max20 and there is no limits issues. all it s going as it should.

maybe contact them. although i doubt they or or they ll ever give u a reset lmao

@MatthewBerman Anthropic lost a lot of money from myself and @CoeusInstitute for their attacks on the open source movement. I don’t mind waiting on the next GLM or gpt model. With a well built harness they do exceptionally well for much less.

@steipete @MatthewBerman we need 5.6 bruv

@MatthewBerman odd.. I'm not having this issue at all

@MatthewBerman ?? did u have no limits or pro account

@odyth @MatthewBerman @LuciferMornens It’s happening to a lot of people. I have Max 20x and hit 5-hr limit almost instantly.

@MatthewBerman They didn’t reset usage too :/

@MatthewBerman Was it the best 5 minutes of your life though?

@moidrk @MatthewBerman It even works when you used up all your quota

@MatthewBerman wasn't it regarding fable 5 or? im a bit confused how you d finish ur limits so fast lol

@MatthewBerman For me- auto switching to 4.8 even for very simple code debug tasks. However, for general discussion- its working fine.

@LuciferMornens Yes, fable

@MatthewBerman Told everyone. •

@MatthewBerman I tried to use all my usage...

@MatthewBerman bah... what's the point. Opus 4.8 and GPT 5.5 got us covered. Not as premium as Fable, but what's the point of having a formula 1 racer if you can only pull it out of the drive way??? and then you gotta push it back into the garage. 🫤

@MatthewBerman Here are the usage results. Looks like everyone gets ~17.5m tokens of usage @ Extra Effort