# @Krubbled — 2026-06-30

♥5 ↻0 · https://x.com/Krubbled/status/2072083759344812058

@theo @anotherdaynow @Zai_org GLM-5.2-max also performs comparably to GPT-5.5-medium on Artificial Analysis for roughly the same cost, so the large price-to-performance difference between models on cursor bench is not necessarily reflective of all benchmarks.

tags: author:krubbled, kind:tweet, model:gpt-5-5, on:glm-5-2, year:2026
cited on: _dossiers/glm-5-2.md, glm-5-2