GLM 4.5 Air
by Z.aitext->text3 endpoints
Neura Intelligence Index
34.4/ 100
Rank
#248—
Confidence
mediumBenchmarks
7(18% coverage)
Overview
GLM-4.5-Air is the lightweight variant of our latest flagship model family, also purpose-built for agent-centric applications. Like GLM-4.5, it adopts the Mixture-of-Experts (MoE) architecture but with a more compact parameter...
Capabilities
Text generation
Tool use / Function calling
Extended reasoning
Prompt caching
Long context
Top-P sampling
Stop sequences
Deterministic seed
Modalities
Input
Text
Output
Text
Technical Specifications
Context Window
131.1K tokens
Max Output
98.3K tokens
Knowledge Cutoff
2024-12-31
Tokenizer
Other
Uptime (24h)
9976.8%
Parameters
106000000000B
License
mitOpen Weights
Country
🇨🇳CN
Release Date
July 28, 20251y
Supported Parameters
frequency_penaltyinclude_reasoningmax_tokenspresence_penaltyreasoningrepetition_penaltyseedstoptemperaturetool_choicetoolstop_ktop_p
Benchmark Performance
coding
| Benchmark | Score | Source |
|---|---|---|
| SWE-bench Verified | 57.6 | verified |
| SciCode | 37.3 | verified |
| TerminalBench | 30 | verified |
reasoning
| Benchmark | Score | Source |
|---|---|---|
| GPQA Diamond | 75 | verified |
| Humanity's Last Exam | 10.6 | verified |
agent
| Benchmark | Score | Source |
|---|---|---|
| TAU-Bench Retail | 77.9 | verified |
| BrowseComp | 21.3 | verified |
Performance Over Time
Pricing
Per 1M tokens
Input$0.13
Output$0.85
Cache read$0.03
Added
July 25, 2025
Last synced 9/21/2026