LLM
How to deploy GLM-4.5-Air on a GPU cloud
A 110B (MoE) language model for chat and instruction-following. Full specs, license and use cases.
GLM-4.5-Air size and hardware requirements
110.5B
Total parameters
Mixture-of-experts: 8 of 128 experts active per token (exact active-parameter count not stated on the model card)
Active parameters
BF16
Published precision
246.9 GB
Min VRAM (native)
| Precision | Weight size on disk | Required VRAM | Cheapest live fit | GPUs needed | Est. $/hr (full fit) |
|---|---|---|---|---|---|
| BF16 | 205.8 GB | 246.9 GB | RTX A6000 | 6 | $1.98/hr |
| FP8 (quantized) | 102.9 GB | 123.5 GB | RTX 4000 SFF Ada | 7 | $1.26/hr |
| INT4 (quantized) | 51.4 GB | 61.7 GB | A100 | 1 | $0.851/hr |
How to run GLM-4.5-Air
Run GLM-4.5-Air with vLLM
Generic example, not from the model's own docs: adjust flags (quantization, context length, parallelism) for your setup.
vllm serve zai-org/GLM-4.5-Air --tensor-parallel-size 6Run GLM-4.5-Air with GGUF quantizations
Prebuilt GGUF weights published at ubergarm/GLM-4.5-Air-GGUF. Run with llama.cpp's llama-server or load the repo directly in LM Studio.
llama-server -hf ubergarm/GLM-4.5-Air-GGUFDeploy GLM-4.5-Air on Aquanode
Aquanode has no one-click deploy template for GLM-4.5-Air; you install the inference engine yourself with the commands below. Aquanode sells GPU pods billed per second, not a hosted inference API.
- Launch a bare GPU pod sized to the requirement above (6× RTX A6000 or larger).
- Open a terminal on the pod, or save one of the commands above as a startup script so it runs automatically the first time the pod boots.
- Run the command and connect to the resulting endpoint.
Launch a GPU pod