AIbase
EN
Home
AI Models Finder
GLM 4.5 AirX
G

GLM 4.5 AirX

GLM-4.5系列的高速推理版本，採用MoE架構，總參數1060億、激活120億，生成速度>100 tokens/s，專為低延遲、高併發場景優化，支持128k上下文與原生工具調用，開源權重
Intelligence(Medium)
Speed(Relatively Slow)
Input Supported Modalities
Yes
Is Reasoning Model
128,000
Context Window
8,192
Maximum Output Tokens
2025-07-01
Knowledge Cutoff
Go Compare

Pricing

￥2.02 /M tokens
Input
￥5.98 /M tokens
Output
￥4.03 /M tokens
Blended Price

Quick Simple Comparison

Input

Output

GLM-4.5
￥0.06
GLM-4.5-AirX
￥0.28
GLM-4.5-Air
￥0.11

Basic Parameters

GLM-4.5-AirXTechnical Parameters
Parameter Count
Not Announced
Context Length
128.00k tokens
Training Data Cutoff
2025-07-01
Open Source Category
Open Source
Multimodal Support
Text, Image
Throughput
100
Release Date
2025-07-28
Response Speed
100 tokens/s

Benchmark Scores

Below is the performance of GLM-4.5-AirX in various standard benchmark tests. These tests evaluate the model's capabilities in different tasks and domains.
Intelligence Index
59.8
Large Language Model Intelligence Level
Coding Index
57.6
Indicator of AI model performance on coding tasks
Math Index
89.4
Capability indicator in solving mathematical problems, mathematical reasoning, or performing math-related tasks
MMLU Pro
81.4
Massive Multitask Multimodal Understanding - Testing understanding of text, images, audio, and video
GPQA
75
Graduate Physics Questions Assessment - Testing advanced physics knowledge with diamond science-level questions
HLE
10.6
The model's comprehensive average score on the Hugging Face Open LLM Leaderboard
LiveCodeBench
70.7
Specific evaluation focused on assessing large language models' ability in real-world code writing and solving programming competition problems
SciCode
37.3
The model's capability in code generation for scientific computing or specific scientific domains
HumanEval
-
Score achieved by the AI model on the specific HumanEval benchmark test set
Math 500 Score
98.1
Score on the first 500 larger, more well-known mathematical benchmark tests
AIME Score
89.4
An indicator measuring an AI model's ability to solve high-difficulty mathematical competition problems (specifically AIME level)
AIbase
Empowering the Future, Your AI Solution Knowledge Base
English简体中文繁體中文にほんご
© 2025AIbase