Search
Zhipu's GLM-4-9B Model Leads Hallucination Rankings with 98.7% Fact Consistency
2025-01-10
Zhipu's GLM-4-9B model has secured the top spot on the HHEM-2.1-Open large model hallucination rankings, achieving an impressive hallucination rate of just 1.3%. This outstanding performance surpasses models from industry leaders such as OpenAI's GPT series and Google's Gemini series. With a fact consistency rate of 98.7% and a response rate of 100%, the GLM-4-9B model stands as the best performer among the 85 evaluated models, demonstrating its exceptional accuracy and reliability.
-
C114 Communication Network -
Communication Home
7 X 24 Track global technological trends
