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.
I'd take this kind of benchmark with a grain of salt. At this point, I have a set of comprehensive guidelines covering both backend and frontend work, and for the frontend we go as far as explaining what we a good design is in our visual system, and even how to conduct a visual review when screenshots are handed to the model.
Deepseek 4.1 ranks very low in this benchmark but it has proven so capable that after being simultaneously on Max x20 and Pro x20 subscriptions, i've transitioned to using DS 4.1 as a daily driver and am very satisfied.
My point is, i think their overall ranking makes sense, matches my experience with out of the box capabilities for vague and underspecified tasks. But seeing a model rank low in their ranking does not mean that the model is incapable. Having skills and guidelines has a lot of influence on what you get out of a model.
I use the direct API from deepseek using Opencode, no problem to report, works like a charm.
Except maybe that the model often believes that he is running out of context, and needs to rush so i occasionally need to jump in to tell it that it still has plenty of room left.
But this does not degrade the quality of my overall experience in a meaningful way.
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.
rednb · · focus · HN ↗
Deepseek 4.1 ranks very low in this benchmark but it has proven so capable that after being simultaneously on Max x20 and Pro x20 subscriptions, i've transitioned to using DS 4.1 as a daily driver and am very satisfied.
My point is, i think their overall ranking makes sense, matches my experience with out of the box capabilities for vague and underspecified tasks. But seeing a model rank low in their ranking does not mean that the model is incapable. Having skills and guidelines has a lot of influence on what you get out of a model.
dotancohen · · focus · HN ↗
I'd love to hear more, I'm considering jumping ship. I'm running a Debian desktop if that's a concern.
rednb · · focus · HN ↗
Except maybe that the model often believes that he is running out of context, and needs to rush so i occasionally need to jump in to tell it that it still has plenty of room left.
But this does not degrade the quality of my overall experience in a meaningful way.
dotancohen · · focus · HN ↗