Model performance at inference in a data center v.s. stripping thinking tokens are effectively the same.
replySure they didn't change the GPUs their running, or the quantization, but if valuable information is removed leading to models performing worse, performance was degraded.
In the same way uptime doesn't care about the incident cause... if you're down you're down no one cares that it was 'technically DNS'.
I thought these days thinking tokens sent my the model (as opposed to used internally) were just for the users benefit. When you send the convo back you have to strip the thinking stuff for next turn. Or is that just local models?
replyClaude code is not infra, the model is the infra. They changed settings to make their models faster and probably cheaper to run too. Honestly with adaptive thinking it no longer matters what model it is if you can dynamically make it do less or more work.
reply