Today's lead story
Opus 5.5 holds the crown for a month — and as Anthropic token use jumps 3.6x, users say the limits are shrinking
First seen on X 13 hours ago@Tech_girl ♥ 3,9487 views
"This model felt endless. First time since launch I've hit usage limits with it." That was developer Dwayne (@CtrlAltDwayne) on X on Oct 25, adding that he thinks Anthropic is "nerfing Opus 5.5 limits." Replies piled on, including one user who said upgrading from the $100 plan to the $200 plan somehow left them with even less usage.
The crowd is real. Citing OpenRouter data, Peter Walker (@PeterJ_Walker) said the 7-day average token volume for Anthropic models is up 3.6x since Opus 5.5 launched. He added in a reply that a high share of that is cache reads — agent loops resend the same context every turn, so token counts can outrun the actual bill.
A post calling Opus 5.5 "the first model in the history of AI to survive an entire month without losing its crown" (@Tech_girl) drew nearly 4,000 likes; the model shipped Sept. 23. AI news watcher Andrew Curran (@AndrewCurran_) wrote that the number of people on his feed switching to Anthropic over the past two weeks "is like nothing I've seen before."
Complaints are stacking up too. Kaleb Campbell (@CampbellKaleb23), who has said since Oct 22 that Claude Code usage drains five times faster, claimed he has confirmation the Claude Code team is digging into the problem. Anthropic has not commented. Separately, Chris (@ChrisGPT) said sources told him Fable 5.5 has been rolled out to enterprise "top clients" such as JPMorgan and Meta, though he said even Anthropic probably doesn't know the public release date.
“For the first time in the history of AI models, a model has survived an entire month without losing its crown. Opus 5.5 needs to be studied.”
Reactions (2)
- @EnjoyerWhoEnjoy♥223
that was on the $100 plan!! I upgraded to the $200 and it’s even less usage - @crsnnn♥7
agent loops on opus resend the same context every turn, so token counts can jump a lot faster than the bill.
