GLM 4.5 Air FP8
GLM 4.5 Air FP8 is a hybrid reasoning model developed by Z.ai, designed for intelligent agent applications. It unifies reasoning, coding, and agent capabilities, offering both a thinking mode for complex tasks and a non-thinking mode for immediate responses.
Input
Output
Context
131K
Max Output
98K
Parameters
110.5B
Technical Specifications
Model TypeChat
Context Window131,072 tokens
Max Output Tokens98,000 tokens
Parameters110.5B
Release DateJul 20, 2025
Training CutoffDec 1, 2024
Licensemit
Open SourceYes
Input Modalities
Text
Output Modalities
Text
Capabilities
Benchmarks
Artificial Analysis
6.8%
HLE
43.7%
AA-LCR
67.3%
AIME
73.3%
GPQA
46.5%
Tau2-Bench
80.7%
AIME 2025
37.6%
IFBench
30.6%
SciCode
96.5%
MATH 500
81.5%
MMLU Pro
80.7
Math
68.4%
LiveCodeBench
16.5
Intelligence
20.5%
Terminal-Bench Hard
Estimated GPU Requirements for GLM 4.5 Air FP8
Estimates based on INT8 quantization. Actual requirements vary by framework and configuration.
More Models by Z.ai
Related Tools
Data sourced from official provider APIs and documentation
Last updated: Aug 8, 2026
Your AI stack shouldn't stand still.
Every month new models become cheaper, faster, and more capable. Inferbase ensures your application automatically benefits without changing a single API call.