Kimi K2.7 and GLM 5.2 are now available in Devin Desktop and CLI. Both models perform well on FrontierCode, our benchmark for real-world engineering tasks graded on mergeability and quality.
FrontierCode ExtendedScore
GLM 5.2 scores 43.0% and Kimi K2.7 scores 39.5% on FrontierCode Extended — placing them in a competitive tier alongside GPT-5.5 (44.8%) and Claude Opus 4.8 (51.8%).
For Pro, Max, and Teams plans, Kimi K2.7 and GLM 5.2 will be free until July 5 — try them now.