supermarket

Language

GLM-4.5-Air

By Z.ai (Zhipu AI), in the GLM family.

Released
28 Jul 2025
Status
Legacy
Context
131.1K
Max output
98.3K
Input per 1M tokens
$0.20
Output per 1M tokens
$1.10
Open weights
Yes
License
mit
Parameters
106B, 12B active
Knowledge cutoff
—
Reasoning
Yes
Modalities
text → text

Category scores

Out of 100, from independent and third-party results only, by method category-2026-09-29.1.

Category scores
CategoryScoreRankCoverage
Coding & software engineering2.9Not ranked1 of 5
Human preference46.2Not ranked1 of 5

Benchmark results

Every result we hold for this model, with who ran it and where it was published.

Knowledge & reasoning

Knowledge & reasoning results
BenchmarkMetricValueProvenanceSettingEvaluatedSource
Humanity's Last Exam (HLE)Accuracy8.1%Self-reported——EvalEval via Hugging Face Community Evals
MMLU-ProAccuracy81.4%Self-reported——EvalEval via Hugging Face Community Evals

Scientific reasoning

Scientific reasoning results
BenchmarkMetricValueProvenanceSettingEvaluatedSource
GPQA DiamondAccuracy71.7%Self-reported——EvalEval via Hugging Face Community Evals

Coding & software engineering

Coding & software engineering results
BenchmarkMetricValueProvenanceSettingEvaluatedSource
GSOOPT@12.9%Third party——GSO Leaderboard via Epoch AI

Human preference

Human preference results
BenchmarkMetricValueProvenanceSettingEvaluatedSource
Text Arena (arena.ai)Arena score (Elo scale)1373Independent—25 Sep 2026Arena
Text Arena (arena.ai)Votes30453Independent—25 Sep 2026Arena

Other GLM versions

Sources