ModelsSep 1, 2026 · 13:148 sources

GLM-5.3 Takes No. 2 Open-Weight Rank on Vals Index to Reach No. 13 Overall

From huggingnews.com

Image: huggingnews.com
AI;DR

The GLM-5.3 AI model reached the No. 13 overall position on the Vals Index with a total score of 57.0, making it the second-highest ranked open-weight model. It earned a 64.3 coding score to become the top open-weight system for programming tasks and secured top spots for legal research and code migration. While performance rose to 78.1% on Vibe Code Bench, a gap remains on difficult Terminal-Bench tasks where GLM-5.3 scored 53.3% compared to Kimi K3's 74.4%.

Read the sourceSign in to suggest

Sources

+ 6 more
  1. SUPPORT@wesroth“78.1% Vibe Code Bench, up 14 points”x.com
  2. SUPPORT@zixuanli_“strong capabilities in legal and financial reasoning”x.com
  3. SUPPORT@dabit3“rolled out support for GLM 5.3 Flash along with an improved model picker UX in Devin CLI”x.com
  4. SUPPORT@factoryai“GLM-5.3 and GLM-5.3 Flash have arrived in Droid”x.com
  5. SUPPORT@factoryai“stronger results on complex programming and long-horizon tasks”x.com
  6. SUPPORT@deeplearningai“This unexpected jump in exploit generation prompted a temporary safety hold on the model weights’ release”x.com

One ranked edition, every morning

The stories that matter today. Free, cancel anytime.

Telegram @aidr_today