GLM-4.6V — benchmark results

Zhipu's open vision-language MoE (106B total/12B active) with native tool calling and 128K context (December 2025). Provider: Zhipu. Released 2025-12-09. Access: Open.

Unified ELO 1593 ± 37, rank #494 of 1776 rated models, from 48 benchmark results.

Strongest benchmark results

BenchmarkScoreMetricPercentile
MechVQA78.91Total (self-reported)93.8
YapBench443.8YapIndex (lower is better)83.5
LithoBench94.5MCQ$_{\mathrm{all}}$ (self-reported)83.3
Ko-AgentBench - L5 Error Handling & Robustness26.01Adaptive Routing Score (%)78.6
MathVision63.5Overall Accuracy (%)76.4
AI for Education Visual Maths - Number and Operations59.46Accuracy (%)75.8
AI for Education Visual Maths - Geometry54.62Accuracy (%)66.7
AI for Education Visual Maths59.92Accuracy (%)65
AI for Education Visual Reasoning - odd one out53.2Accuracy (%)62.9
AI for Education Visual Maths - Algebra76.92Accuracy (%)60.8
Chatbot Arena (Text)1377Elo58.8
AI for Education Visual Maths - Measurement75.68Accuracy (%)58.3

Interactive version: theaggregate.ai/model?slug=glm-4-6v · How the rankings work · Data refreshed daily, snapshot 2026-07-22.