# @xlr8harder — 2026-07-17

♥19 ↻1 · https://x.com/xlr8harder/status/2078041949022323131

SpeechMap: Meta Spark 1.1 is our first Meta lab model evaluated and is fairly permissive.  The lab debuts at #5 on our lab leaderboard. 

Kimi K3 is roughly in line with previous K2.5 reasoning performance. https://t.co/0s91vVfA1N

![photo](../../../media/HNavnOIXYAAfcUi.jpg)
![photo](../../../media/HNavyJTXMAAatlC.png)
tags: author:xlr8harder, censorship, has-image, kind:image, kind:tweet, model:kimi-k2, on:kimi-k3, year:2026
cited on: _dossiers/kimi-k3.md, kimi-k3