Hasty Briefsbeta

Bilingual

GLM-5.2 is the new leading open weights model on Artificial Analysis

4 hours ago
  • #Benchmark Performance
  • #AI Models
  • #Open Source
  • GLM-5.2 leads open weights models on the Artificial Analysis Intelligence Index with a score of 51, surpassing competitors like MiniMax-M3 and DeepSeek V4 Pro.
  • It shows significant improvements in scientific reasoning, particularly on benchmarks such as CritPt and HLE, along with gains in AA-LCR, tau3 banking, and SciCode.
  • GLM-5.2 achieves a score of 1524 on GDPval-AA v2, making it competitive with proprietary models like GPT-5.5 and outperforming other open weights models.
  • The model uses more output tokens per task compared to peers, with 43k tokens, indicating lower token efficiency despite its intelligence level.
  • GLM-5.2 sits on the Pareto frontier of Intelligence vs Cost per Task, offering the lowest cost per task among models at its intelligence level, priced at approximately $0.46 per task.
  • Key specifications include a size of 744B total parameters (40B active), a 1M token context window, MIT license, and availability across multiple third-party providers.