ModelsAug 30, 2026 · 09:354 sources

GLM-5.3 Beats GPT-5.6 Sol on Terminal Bench 4.0 in Open Weights Win Over Frontier Model

From huggingnews.com

AI;DR

GLM-5.3 scored 41.82% on the Terminal-Bench 4.0 agentic software engineering rankings, surpassing GPT-5.6 Sol's score of 37.27%. Opus 5 and Fable 5 took the top two spots with scores of 51.82% and 44.55%, respectively. The updated benchmark incorporates user feedback and reduces the total tasks from 74 to 66 to reduce noise from infrastructure failures.

Read the sourceSign in to suggest

Sources

+ 2 more
  1. SUPPORT@cwolferesearch“Terminal Bench 4 also marks the transition of Terminal Bench to a continuously evolving benchmark”x.com
  2. SUPPORT@teksedge“GLM-5.3's reported benchmark cost was also much lower than Fable”x.com

One ranked edition, every morning

The stories that matter today. Free, cancel anytime.

Telegram @aidr_today