14:06
2026-07-16
runtimewire.com
artificial-intelligence
Head to head: Kimi-K2.7-Code vs gpt-5.4
In a head-to-head benchmark of 12 fresh text tasks, Kimi-K2.7-Code scored 103.5 against gpt-5.4's 101.5, with only 50% confidence that either model is better. The results show a near tie, with Kimi sl…