# @maxbittker — 2026-07-17

♥70 ↻4 · https://x.com/maxbittker/status/2078207968185581647

Kimi K3 did poorly on Runescape bench because it thinks a lot and has slow inference. Performs worse than Kimi K2.7, way worse than GLM 5.2, maybe a different story when better providers show up. https://t.co/OBXD6DqhzI

![photo](../../../media/HNdFvNhXYAA7raD.jpg)
tags: author:maxbittker, has-image, kind:image, kind:tweet, model:kimi-k2, on:glm-5-2, on:kimi-k2-7-code, on:kimi-k3, year:2026
cited on: _dossiers/glm-5-2.md, _dossiers/kimi-k3.md, glm-5-2, kimi-k2-7-code, kimi-k3