[Bug] SEVERELY Degraded Opus 4.8 Preformance, Severe Lags and Inadequate Reasoning

Bug Description
Opus 4.8 has been nerfed into the ground and I'm done being quiet about it.

I'm a paying Max 20x customer and I am genuinely disgusted by what you're doing to this model. Opus 4.8 has been gutted over the last few days. Crank the reasoning up to extra-high and it crawls — it sits there thinking for over ten minutes before it gives me anything. Ten minutes. For one response.

And when it's not grinding for ten minutes, it's brain-dead. The model is severely degraded compared to where it was just a couple of days ago. It can't do even the simplest tasks. It doesn't know what it is. It ignores instructions and rules completely — every single rule, gone. This is not the model I was using days ago.

So pick your poison: unbearably slow, or unbearably stupid. There is no setting where it works like it used to.

This is criminal. You are silently degrading a product people pay top dollar for, and you do it over and over again. I pay for the highest tier you offer and you quietly take away what I'm paying for, with no warning, no acknowledgement, nothing. That is customer abuse, plain and simple, and it is a repeated, consistent pattern from Anthropic.

I'm shocked at the negligence. _Stop nerfing the models. Restore Opus 4.8 to what it was, and be honest with the people paying you when you change it._

Environment Info

• Platform: darwin
• Terminal: ghostty
• Version: 2.1.185
• Feedback ID: dd3d56c0-11c3-4611-b4d3-dcd97cf37b1e

Errors

[]