ninjahawk/livenerf

Benchmark for tracking model capability after release.

926 best points · 396 comments · 4 Hacker News posts · 1,725 GitHub stars · View on GitHub

Hacker News discussions

• Livenerf: Has Opus 5.5 been nerfed yet? (926 points, 394 comments, by bryan0, 2026-09-29)
• Show HN: Livenerf – a benchmark for whether Opus 5.5 gets nerfed (5 points, 2 comments, by ninjahawk1, 2026-09-22)
• Show HN: Has Anthropic been nerfing their models without disclosure (4 points, 0 comments, by ninjahawk1, 2026-09-27)
• LiveNerf Adopts MIT License (2 points, 0 comments, by ninjahawk1, 2026-10-01)