GLM-4.5-Air

GLM-4.5-Air is a more compact variant of GLM-4.5 designed for efficient Agentic, Reasoning, and Coding (ARC) applications. It features 106 billion total parameters with 12 billion active parameters using MoE architecture. Like GLM-4.5, it is a hybrid reasoning model providing thinking mode for complex reasoning and tool usage, and non-thinking mode for immediate responses. Despite its compact design, GLM-4.5-Air delivers competitive performance with a score of 59.8 across 12 industry-standard benchmarks, ranking 6th overall while maintaining superior efficiency. It supports 128K context length and is released under MIT open-source license allowing commercial use.

Benchmark results

Benchmark Score Tags Source
AA-Index 64.8% self-reported llm-stats link →
AIME 2024 89.4% self-reported llm-stats link →
BFCL-v3 76.4% self-reported llm-stats link →
BrowseComp 21.3% self-reported llm-stats link →
GPQA 75.0% self-reported llm-stats link →
HLE 10.6% self-reported llm-stats link →
Humanity's Last Exam 10.6% self-reported llm-stats link →
LiveCodeBench 70.7% self-reported llm-stats link →
MATH-500 98.1% self-reported llm-stats link →
MMLU-Pro 81.4% self-reported llm-stats link →
SciCode 37.3% self-reported llm-stats link →
SWE-Bench Verified 57.6% self-reported llm-stats link →
TAU-bench Airline 60.8% self-reported llm-stats link →
TAU-bench Retail 77.9% self-reported llm-stats link →
Terminal-Bench 30.0% self-reported llm-stats link →