glm-4-9B Benchmark Results

glm-4-9B — benchmark results

Provider: Zhipu. Released 2024-06-04. Access: Open.

Unified ELO 1370 ± 51, rank #1342 of 1776 rated models, from 23 benchmark results.

Benchmark	Score	Metric	Percentile
Benchmarking Large Language Models for Graphem	57.56	Direct (self-reported)	86.1
Open PL LLM - RAG	71.01	Average RAG Score (%)	80.4
Open LLM Leaderboard - MuSR	14.19	Score	77.6
Open LLM Leaderboard - MMLU-Pro	34.94	Score	75.5
Open LLM Leaderboard - BBH	35.81	Score	72.7
Open PL LLM - Generative	60.68	Average Generative Score (%)	72.7
Open LLM Leaderboard - GPQA	8.84	Score	72.5
Open PL LLM Leaderboard	55.32	Average Score (%)	69.2
Greek MMLU - STEM	63.19	Accuracy (%)	66.7
Greek MMLU	64.98	Accuracy (%)	65.4
Greek MMLU - Other	63.41	Accuracy (%)	65.4
Open PL LLM - Multiple Choice	48.34	Average Multiple-Choice Score (%)	64.7

Interactive version: theaggregate.ai/model?slug=glm-4-9b · How the rankings work · Data refreshed daily, snapshot 2026-07-22.