What GPU do I need to run zai-org/GLM-4.5V?
A 108B vision-language model that reads images alongside text. 107.7B parameters, published in BF16. View on Hugging Face
GLM-4.5V is published by zai-org on Hugging Face, with 41,093 downloads and 722 likes to date. It's a Glm4vMoeForConditionalGeneration model built for image-text-to-text, published natively in BF16.
What GLM-4.5V is
GLM-4.5V is a 108B-parameter vision-language model published by Z.ai (Zhipu) on Hugging Face. It is released under MIT.
License note: permissive: allows commercial use, modification and redistribution. Facts in this section are sourced from GLM-4.5V's Hugging Face model card, not benchmarked by Aquanode.
What it's used for
- Image understanding and captioning
- Visual question answering
- Document/OCR-style reading
VRAM required & cheapest live GPU fit
Required VRAM = weight size at each precision, plus a fixed overhead for KV-cache, activations, and fragmentation. Full formula and assumptions: methodology.
| Precision | Weight size | Required VRAM | Cheapest live fit | GPUs needed | Est. $/hr (full fit) |
|---|---|---|---|---|---|
| BF16 | 200.6 GB | 240.8 GB | RTX A6000 | 6 | $2.18/hr |
| FP8 (quantized) | 100.3 GB | 120.4 GB | RTX 5060 Ti | 8 | $0.880/hr |
| INT4 (quantized) | 50.2 GB | 60.2 GB | A100 | 1 | $1.21/hr |
| cheaper alt. | RTX 5060 Ti | 4 | $0.440/hr |
A GPU is only matched to a row if its hardware supports that precision, and the primary recommendation is always a single-GPU fit when one exists.
INT4 requires a quantized checkpoint actually published for this model, check its Hugging Face page before relying on this row.
Cheapest way to run GLM-4.5V at its published (BF16) precision: 6× RTX A6000, at $0.363/hr per GPU ($2.18/hr total). Quantizing to FP8 or INT4 (rows above) can cost less, but requires a compatible quantized checkpoint to exist for this model.
GLM-4.5V: common questions
Can GLM-4.5V run on a single GPU?
No. At BF16 it needs 240.8 GB of VRAM, more than a single 32 GB desktop card holds. The cheapest capable card in the live feed is a 48.0 GB RTX A6000, and it takes 6 of them.
How many GPUs do I need to run GLM-4.5V?
6 at BF16. It needs 240.8 GB of VRAM and the cheapest capable live offer is a 48.0 GB RTX A6000, so 6 of them come to $2.18/hr in total.
Does quantizing GLM-4.5V lower the GPU bill?
Yes. At BF16 the cheapest live fit is 6 RTX A6000 cards at $2.18/hr. At FP8 (quantized) it drops to 8 RTX 5060 Ti cards at $0.880/hr, provided a quantized checkpoint exists for it.
How to run GLM-4.5V
Run GLM-4.5V with vLLM
From zai-org/GLM-4.5V's own deployment docs.
vllm serve zai-org/GLM-4.5V \
--tensor-parallel-size 4 \
--tool-call-parser glm45 \
--reasoning-parser glm45 \
--enable-auto-tool-choice \
--served-model-name glm-4.5v \
--allowed-local-media-path / \
--media-io-kwargs '{"video": {"num_frames": -1}}'Source: https://huggingface.co/zai-org/GLM-4.5V/raw/main/README.md
Run GLM-4.5V with GGUF quantizations
Prebuilt GGUF weights published at mradermacher/GLM-4.5V-GGUF. Run with llama.cpp's llama-server or load the repo directly in LM Studio.
llama-server -hf mradermacher/GLM-4.5V-GGUFDeploy GLM-4.5V on Aquanode
Aquanode has no one-click deploy template for GLM-4.5V; 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 podWeight-to-VRAM math, the fit rules, and how live prices are normalized: full methodology.
More GLM-4.5 models
- GLM-4.5-Air (110.5B, BF16)
- GLM-4.5-Air (110.5B, BF16)
- GLM-4.5-Air-Base (110.5B, BF16)
- GLM-4.5-Air-FP8 (110.5B, F8_E4M3)
- GLM-4.6 (356.8B, BF16)
Related reading: RTX A6000 pricing and specs, The NVIDIA Inception program, explained, Google Colab alternatives for dedicated GPU access, Free GPU credits for students and researchers, and RunPod volume disk vs. network volume, compared.
