Zhipu AI/Mixture of Experts

GLM 4.5 Air

Name: GLM 4.5 Air
Author: Zhipu AI

GLM 4.5 Air — lighter MoE variant balancing speed and capability.

chatcodingreasoningtool_useThinking

110B

Parameters (12B active)

125K

Context length

Benchmarks

Quantizations

378K

HF downloads

Architecture

MoE

Released

2025-08-08

Layers

KV Heads

Head Dim

128

Family

glm

Quantization Options

Quant	Bits	VRAM	Quality
IQ2_XXS	2.38	33.2 GB	low
IQ2_M	2.93	40.8 GB	low
Q2_K	3.16	43.9 GB	low
IQ3_XXS	3.25	45.2 GB	low
IQ3_XS	3.5	48.6 GB	low
Q3_K_S	3.64	50.5 GB	low
IQ3_M	3.76	52.2 GB	low
Q3_K_M	4	55.5 GB	low
Q3_K_L	4.3	59.6 GB	moderate
IQ4_XS	4.46	61.8 GB	moderate
Q4_K_S	4.67	64.7 GB	moderate
Q4_K_M	4.89	67.7 GB	good
Q5_K_S	5.57	77.1 GB	good
Q5_K_M	5.7	78.9 GB	good
Q6_K	6.56	90.7 GB	excellent
Q8_0	8.5	117.4 GB	lossless
FP16	16	220.5 GB	lossless

Select your GPU above to see speed estimates and compatibility for each quantization.

▸ READY TO RUN THIS?RENT BY THE HOUR

RENT A GPU AND RUN GLM 4.5 AIR NOW

Rent on RunPod →Or Vast.ai →

Spin up an A100 / H100 / 4090 in ~60s. Pay by the second. Cancel anytime.

Community Ratings

Loading ratings...

Benchmarks (14)

MATH-50096.5

MMLU-PRO81.5

AIME80.7

AA Math80.7

GPQA Diamond73.3

LiveCodeBench68.4

τ²-Bench46.5

AA Long Context43.7

IFBench37.6

SciCode30.6

AA Coding23.8

AA Intelligence23.2

Terminal-Bench20.5

HLE6.8

Run this model

▸Easiest way to get started·Beginners

DOCS ↗

curl -fsSL https://ollama.com/install.sh | sh

$ollama run glm:110b-q4_K_M

Tag may need adjustment — check ollama.com/library/glm for available tags.

▸ SETUP GUIDE

Auto-setup with fitmyllm CLI

Detects your GPU, recommends the best model, downloads it, and starts chatting — zero config. Benchmarks your speed and contributes anonymous data to improve predictions.

pip install fitmyllmthen run fitmyllmLearn more

Auto-detect GPULive tok/s in chatSpeed benchmarks9 inference engines

HuggingFace GGUF Downloads Build Hardware