GLM-4.5-Air
Zero-eval
#1TAU-bench Airline
#2MATH-500
#2BFCL-v3
+3 more
by Zhipu AI
About
GLM-4.5-Air is a language model developed by Zhipu AI. It achieves strong performance with an average score of 60.8% across 14 benchmarks. It excels particularly in MATH-500 (98.1%), AIME 2024 (89.4%), MMLU-Pro (81.4%). It's licensed for commercial use, making it suitable for enterprise applications. Released in 2025, it represents Zhipu AI's latest advancement in AI technology.
Timeline
AnnouncedJul 28, 2025
ReleasedJul 28, 2025
Specifications
License & Family
License
MIT
Benchmark Performance Overview
Performance metrics and category breakdown
Overall Performance
14 benchmarks
Average Score
60.8%
Best Score
98.1%
High Performers (80%+)
3Top Categories
math
98.1%
general
76.0%
agents
53.3%
code
46.0%
reasoning
37.7%
Additional Information
Content coming soon...
All Benchmark Results for GLM-4.5-Air
Complete list of benchmark scores with detailed information
| MATH-500 | math | text | 0.98 | 98.1% | Self-reported |
| AIME 2024 | general | text | 0.89 | 89.4% | Self-reported |
| MMLU-Pro | general | text | 0.81 | 81.4% | Self-reported |
| TAU-bench Retail | agents | text | 0.78 | 77.9% | Self-reported |
| BFCL-v3 | general | text | 0.76 | 76.4% | Self-reported |
| GPQA | general | text | 0.75 | 75.0% | Self-reported |
| LiveCodeBench | code | text | 0.71 | 70.7% | Self-reported |
| AA-Index | reasoning | text | 0.65 | 64.8% | Self-reported |
| TAU-bench Airline | agents | text | 0.61 | 60.8% | Self-reported |
| SWE-Bench Verified | general | text | 0.58 | 57.6% | Self-reported |
Showing 1 to 10 of 14 benchmarks