livenerf tracks Opus 5.5 performance drift after launch

On 2026-09-30, Hacker News' front page featured livenerf, an open-source, append-only benchmark that checks whether a model gets worse after release. It currently targets Claude Opus 5.5. The benchmark runs once a day for 30 days starting 2026-09-24. So far only the project's launch and run schedule are known; the report gives no drift results.

Written by AI from the coverage · updated 52 minutes ago

Not enough continuous observations to draw a trend yet.