Z.ai: GLM 4.5 Air

Z.ai: GLM 4.5 Air

z-ai · Released Jul 25, 2025 Professional
Intelligence #10 / 565
82.0 Our Score
Speed #144 / 262
89.5 tokens / sec
Input #228 / 565
$0.125 per 1M tokens
Output #301 / 565
$0.850 per 1M tokens
Context #228 / 565
131,072 tokens

Analysis Summary

Z.ai: GLM 4.5 Air sits in the Professional tier on our leaderboard, ranked #10 of 565 published models on overall intelligence. At $0.125 input and $0.850 output per 1M tokens, it is among the most expensive on the market. It offers a standard large context window and supports tool use, function calling, and reasoning.

Editorial notes

Z.ai GLM 4.5 Air delivers strong coding and agentic benchmarks at low cost with tool use and function calling; math scores are high and the price-performance ratio is competitive, though provider adoption outside Asia is limited.

Assessed May 31, 2026

Rankings consider pricing, capabilities, benchmarks, and real-world applicability and are refreshed as new models launch. Feedback?

Performance Profile

Intelligence4Technical4.4Value7.8Content4.7
Intelligence 4/10
Technical 4.4/10
Content 4.7/10
Value 7.8/10

How Z.ai: GLM 4.5 Air compares

Z.ai: GLM 4.5 Air ranks #148 of 370 AI models we track for overall intelligence, #119 of 307 for coding, #125 of 282 for agentic tasks. Its 131K-token context window is larger than 60% of the models we list. At $0.13 per million input tokens it is cheaper than 60% of comparable models.

About Z.ai: GLM 4.5 Air

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

Tool Use Function Calling

Performance Indices

Source: Artificial Analysis

23.2 Intelligence Index
23.8 Coding Index
33.5 Agentic Index
80.7 Math Index

Benchmark Scores

Intelligence

GPQA Diamond 73.3% Graduate-level scientific reasoning
HLE 6.8% Humanity's Last Exam
MMLU Pro 81.5% Multi-task language understanding
MATH 500 96.5% Mathematical problem-solving
AIME 67.3% Competition mathematics
AIME 2025 80.7% Competition mathematics (2025)
SciCode 30.6% Scientific computing

Technical

LiveCodeBench 68.4% Live coding evaluation
TerminalBench Hard 20.5% Agentic terminal tasks
τ²-Bench 46.5% Conversational agent benchmark

Content

IFBench 37.6% Instruction following
LCR 43.7% Long-context reasoning

Benchmark data from Artificial Analysis and Hugging Face

How does Z.ai: GLM 4.5 Air stack up?

Compare side-by-side with other professional models.

Compare Models

Model Information

OpenRouter ID z-ai/glm-4.5-air
Providerz-ai
Release Date July 25, 2025
Context Length131,072 tokens
Max Completion131,070 tokens
Status Active

Pricing

Token Type Cost per 1M tokens Cost per 1K tokens
Input $0.13 $0.000125
Output $0.85 $0.000850

Live Performance

Live endpoint metrics, refreshed every 30 minutes.

99.7%
Avg Uptime
853ms
Best Latency (TTFT)
33 tok/s
Best Throughput
4/4
Active Endpoints
Available via: Io Net, Novita, SiliconFlow, Z.AI

Leaderboard Categories

Frequently asked questions about Z.ai: GLM 4.5 Air

How much does Z.ai: GLM 4.5 Air cost?

Z.ai: GLM 4.5 Air costs $0.13 per million input tokens and $0.85 per million output tokens.

What is the context window of Z.ai: GLM 4.5 Air?

Z.ai: GLM 4.5 Air has a context window of 131,072 tokens (131K).

Is Z.ai: GLM 4.5 Air good for coding?

On our coding benchmark index, Z.ai: GLM 4.5 Air ranks #119 of 307 models, placing it in the broader range of the field for code generation and debugging.

What can Z.ai: GLM 4.5 Air do?

Z.ai: GLM 4.5 Air supports tool use and function calling.

Who created Z.ai: GLM 4.5 Air?

Z.ai: GLM 4.5 Air is developed by Z.ai and was released on July 25, 2025.