What GPU do I need to run zai-org/GLM-4.5-Air-Base?
110.5B parameters, published in BF16. View on Hugging Face
Set up GLM-4.5-Air-BaseGLM-4.5-Air-Base is published by zai-org on Hugging Face, with 10,916 downloads and 46 likes to date. It's a Glm4MoeForCausalLM model built for text-generation, published natively in BF16.
VRAM required & cheapest live GPU fit
Required VRAM = weight size at each precision times a flat 1.2 overhead for activations and fragmentation. The KV-cache grows with context and is not in that factor; it is listed per model below where the architecture is published. Full formula and assumptions: methodology.
| Precision | Weight size | Required VRAM | Cheapest live fit | GPUs needed | Est. $/hr (full fit) |
|---|---|---|---|---|---|
| BF16 | 205.8 GB | 246.9 GB | RTX A6000 | 6 | $2.18/hr |
| FP8 (quantized) | 102.9 GB | 123.5 GB | AMD MI300X | 1 | $2.63/hr |
| cheaper alt. | RTX 4000 SFF Ada | 7 | $1.39/hr | ||
| INT4 (quantized) | 51.4 GB | 61.7 GB | A100 | 1 | $1.16/hr |
| cheaper alt. | RTX A5000 | 3 | $0.528/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-4.5-Air-Base 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-4.5-Air-Base: KV cache by context length
The KV cache is the memory the attention layers hold for every token of context, on top of the weights. It is not part of the flat 1.2x overhead in the table above, grows with context length and with every concurrent request, and is why a long-context deployment needs more VRAM than the table shows.
Grouped-query attention: every layer caches keys and values for a few shared KV heads. Cached per token: 188,416 bytes at 16-bit.
| Context | KV cache, one sequence |
|---|---|
| 4K | 0.72 GB |
| 32K | 5.8 GB |
| 128K | 23.0 GB |
Computed from the layer, head and window counts in the model's own configuration (zai-org/GLM-4.5-Air-Base), at 16 bits per cached value. Each concurrent sequence needs its own cache, on top of the weights.
GLM-4.5-Air-Base: common questions
Can GLM-4.5-Air-Base run on a single GPU?
No. At BF16 it needs 246.9 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-4.5-Air-Base?
6 at BF16. It needs 246.9 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-4.5-Air-Base lower the GPU bill?
Yes. At BF16 the cheapest live fit is 6 RTX A6000 cards at $2.18/hr. At INT4 (quantized) it drops to one A100 at $1.16/hr, provided a quantized checkpoint exists for it.
Weight-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-FP8 (110.5B, F8_E4M3)
- GLM-4.6 (356.8B, BF16)
- GLM-4.5 (358.3B, BF16)
- GLM-4.7 (358.3B, BF16)
Lineage
Quantized and fine-tuned versions
Alternatives at this size
Other models for text-generation within about a third of GLM-4.5-Air-Base's 110.5B parameters, from other model lines.
- NVIDIA-Nemotron-3-Super-120B-A12B-BF16 (123.6B, BF16)
- Qwen3-Coder-Next (79.7B, BF16)
- Laguna-S-2.1 (117.6B, BF16)
- Hunyuan-A13B-Instruct (80.4B, BF16)
- Ling-3.0-flash (127.5B, BF16)
More on GLM-4.5-Air-Base
Fits on a 288 GB GPU at BF16: every model that fits in 288 GB.
Fits on a 141 GB GPU at FP8: every model that fits in 141 GB.
Fits on an 80 GB GPU at INT4: every model that fits in 80 GB.
Compare GLM-4.5-Air-Base with NVIDIA-Nemotron-3-Super-120B-A12B-BF16, Qwen3-Next-80B-A3B-Instruct, Hunyuan-A13B-Instruct, Qwen3-Next-80B-A3B-Thinking, NVIDIA-Nemotron-3-Super-120B-A12B-Base-BF16.
Related reading: RTX A6000 pricing and specs, How much VRAM you need for LLMs, Serving LLMs with vLLM, vLLM vs TensorRT-LLM vs SGLang, and Best GPU for LLM inference.
