About
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...
Related Models
Pricing
Input
$0.13
per 1M tokens
Output
$0.85
per 1M tokens
Blended
$0.31
per 1M tokens
Cheaper than 61% of models. Median price is $0.54/1M tokens.
Cost Calculator
Tokens per day1M
100K100M
Daily
$0.31
Monthly
$9.30
vs. Similar Models
Grok 4 Fast (Non-reasoning)Q:0.0
$0.28-11%
GPT-5.4 mini (Non-Reasoning)Q:+0.1
$1.69+445%
Nova 2.0 Omni (low)Q:+0.1
$0.85+174%
OpenAI: GPT-4.1 MiniQ:-0.2
$0.70+126%
Performance
75
tokens/sec
Faster than 37% of models
1.49
seconds
Faster than 34% of models
28.06
seconds
Faster than 17% of models
Market Median
94 tok/s
20% slower
Median TTFT
1.10s
35% slower
Throughput/Dollar
243
tok/s per $/1M
Speed Comparison
NVIDIA Nemotron Nano 9B V2 (Reasoning)
75 tok/s+0%
MiniMax: MiniMax M3
75 tok/s+0%
DeepSeek: DeepSeek V4 Pro
75 tok/s+0%
Context Window
131K
tokens
Larger than 27% of models
Max Output
98K
tokens
75% of context
Benchmarks
MMLU-Pro
81.5%
GPQA Diamond
73.3%
HLE
6.8%
LiveCodeBench
68.4%
SciCode
30.6%
TerminalBench Hard
20.5%
MATH-500
96.5%
AIME
67.3%
AIME 2025
80.7%
IFBench
37.6%
Long Context Recall
43.7%
Tau2
46.5%
Market AverageTop Score
