On: Livenerf: Has Opus 5.5 been nerfed yet?

[scrollop]

There's also this one which has been around for a while

https://marginlab.ai/trackers/claude-code/

[sscaryterry → scrollop]

The tracker for Codex resonates with me. Its thick as pig shit the last few days: https://marginlab.ai/trackers/codex/

[phoghed → sscaryterry]

Seems like a harness change (maybe bug) rather than a model change, the input tokens dropped quite a bit right when the degradation happened.

[lxgr → sscaryterry]

> We use the latest available Codex release with GPT-6 Sol.

This alone makes the benchmark unsound.

[shawabawa3 → sscaryterry]

> "We are collecting a new GPT-6 Sol/high baseline from runs beginning September 24, 2026. Degradation detection is paused."