Dev Questions DeepSeek 0731 Benchmark Superiority Over GLM 5.2

Sentdex · x · 2026-08-06

Developer Sentdex expressed skepticism on X regarding recent benchmark scores, stating that he does not feel DeepSeek's latest version (DSV4F-0731) is equivalent or better than GLM 5.2 based on personal use. He challenged the community to independently verify the model's scores using a consistent testing harness.

Related event: Developer Questions DeepSeek New Model Benchmark After Testing(2 posts)→

Original post →

More from Models

Models channel →