GLM-4.5-Air (Non-reasoning): benchmark results

GLM 4.5 Air evaluated with reasoning disabled. Provider: Zhipu. Released 2025-07-28. Access: Open.

Unified ELO 1529 ± 1, rank #1134 of 3078 rated models, from 150 benchmark results.

Strongest benchmark results

BenchmarkScoreMetricPercentile
EuroEval Finnish NLU - ScaLA FI41.61Linguistic acceptability Score (%)92.2
EuroEval Spanish NLU - ScaLA ES40.78Linguistic acceptability Score (%)91.2
EuroEval Swedish NLU - Swerec78.96Sentiment classification Score (%)90
EuroEval English Knowledge92.31Knowledge Average Score (%)89.1
EuroEval Italian NLU - Sentipolc1665.51Sentiment classification Score (%)88.7
EuroEval Norwegian NLU - ScaLA NN50.85Linguistic acceptability Score (%)87.6
EuroEval Portuguese NLU - SST-2 PT83.84Sentiment classification Score (%)86.8
EuroEval German NLU - ScaLA DE51.28Linguistic acceptability Score (%)86.7
EuroEval Spanish NLU - Sentiment Headlines ES49.06Sentiment classification Score (%)86.7
UGI - Writing47Writing Score86.7
EuroEval Norwegian NLU - ScaLA NB62.43Linguistic acceptability Score (%)85.4
EuroEval Norwegian Knowledge - Idioms NO33.77MCC (x100)84.9

Interactive version: theaggregate.ai/model?slug=glm-4-5-air-non-reasoning · How It Works · Data refreshed daily, snapshot 2026-09-19.