GLM-4.5-Air
Z AIGLMOpen WeightMIT · Commercial OK
Description
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.
Release Date
2025-07-28
Parameters
106.0B
Context Length
131K
Modalities
text
Capability Radar
35
general
60
coding
79
reasoning
45
science
70
agents
0
multimodal
Rankings
| Domain | #Rank | Score | Source |
|---|---|---|---|
| Agentic Capability | 86 | 42.0 | LS |
| Code Ranking | 203 | 54.0 | AA |
| General Ranking | 291 | 45.0 | AA |
| Science | 292 | 46.0 | AA |
Benchmark Scores (LLM Stats)
(LLM Stats (zeroeval))Agents
BFCL-v3
76.4%SR
Terminal-Bench
30.0%SR
BrowseCompOpenAI (2025)
21.3%SR
Biology
GPQANYU + Cohere + Anthropic (2023)
75.0%SR
SciCode
37.3%SR
Code
LiveCodeBench
70.7%SR
SWE-Bench Verified
57.6%SR
Communication
TAU-bench Retail
77.9%SR
TAU-bench Airline
60.8%SR
Finance
MMLU-Pro
81.4%SR
General
AA-Index
64.8%SR
Math
MATH-500
98.1%SR
AIME 2024
89.4%SR
Humanity's Last Exam
10.6%SR
AA Evaluation Indices
(Artificial Analysis)Math Index(Artificial Analysis)80.7
Intelligence Index(Artificial Analysis)16.7
Math 500(OpenAI (2024), subset of Hendrycks et al. MATH (2021))1.0
Mmlu Pro(TIGER-Lab (Univ. of Waterloo, Toronto, CMU, 2024))0.8
Aime 25(MAA (Mathematical Association of America))0.8
Gpqa(NYU + Cohere + Anthropic (2023))0.7
Livecodebench(UC Berkeley + MIT + Cornell (2024))0.7
Aime(MAA (Mathematical Association of America))0.7
Tau2(Sierra + U Toronto + Vector Institute (2025))0.5
Lcr(Artificial Analysis)0.5
Ifbench(Google Research (2023))0.4
Scicode(UIUC + Argonne National Lab (2024))0.3
Terminalbench Hard(Stanford × Laude Institute (2026))0.2
Hle(Center for AI Safety + Scale AI (2025))0.1
LLM Stats Category Scores
(LLM Stats (zeroeval))Legal80
Structured Output80
Language80
Finance80
Healthcare80
Communication70
Tool Calling70
Math60
Physics60
Reasoning60
Frontend Development60
General60
Biology60
Chemistry60
Code50
Agents40
Search20
Vision10
Pricing
Input Price$0.17 / 1M tokens
Output Price$0.98 / 1M tokens
Blended Price (3:1)$0.372 / 1M tokens
Cache Read Price$0.03 / 1M tokens
Cache Write PriceFree
Speed
Tokens/sec0.0
Time to First Token0.00s
Time to Answer0.00s
Provider Price Ranking
Provider Price Ranking
9 providers
Cheapest: ZenMuxMost Expensive: OrcaRouter
ProviderInputOutput
1ZenMuxCheapest
$0.11
$0.56
2302.AI
$0.1143
$0.286
3OpenRouter
$0.13
$0.85
4Kilo Gateway
$0.13
$0.85
5LLM Gateway
$0.13
$0.85
6Z AIPRIMARY
$0.17
$0.98
7Z.AI
$0.2
$1.1
8Zhipu AI
$0.2
$1.1
9OrcaRouter
$0.2
$1.1
Compare pricing across different API providers for this model.