They test it on launch day, then benchmark it against that. A deviation of above 10% is considered a change. They're currently tracking Opus 5.5 and GPT-6 Astra.
This bench famously detected a degradation of Opus 4.6 which Anthropic later blogged about. I personally think people sense nerfs more often than they happen and that it's often about honeymoon effects.
Theory (Conjecture? Hypothesis?): What we notice as "model nerfing" is the company diverting compute to training/running new unreleased models..
Remember that some people get access to the next flagship version long before us peasants do. I recall seeing the mention of "Astra" more than a month before it was officially announced
The more likely thing that would happen is that the provider begins silently interpreting (perhaps some) high effort-level requests as medium, etc., or having a classifier do this far more subtly. As such, the load on the cluster is less, and more resources can be devoted to training. Whether the frontier labs actually do this is purely conjecture at this point.
jug · · focus · HN ↗
<a href="https://www.bridgebench.ai/nerf-bench" rel="nofollow">https://www.bridgebench.ai/nerf-bench
They test it on launch day, then benchmark it against that. A deviation of above 10% is considered a change. They're currently tracking Opus 5.5 and GPT-6 Astra.
This bench famously detected a degradation of Opus 4.6 which Anthropic later blogged about. I personally think people sense nerfs more often than they happen and that it's often about honeymoon effects.
Razengan · · focus · HN ↗
Remember that some people get access to the next flagship version long before us peasants do. I recall seeing the mention of "Astra" more than a month before it was officially announced
Centigonal · · focus · HN ↗
btown · · focus · HN ↗
nightpool · · focus · HN ↗
zxilly · · focus · HN ↗