What GPU do I need to run zai-org/GLM-4.5-Air-FP8?
110.5B parameters, published in F8_E4M3. View on Hugging Face
Set up GLM-4.5-Air-FP8GLM-4.5-Air-FP8 is published by zai-org on Hugging Face, with 83,933 downloads and 82 likes to date. It's a Glm4MoeForCausalLM model built for text-generation, published natively in F8_E4M3.
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) |
|---|---|---|---|---|---|
| FP8 (native) | 102.9 GB | 123.5 GB | RTX 4090 | 3 | $1.32/hr |
| INT4 (quantized) | 51.5 GB | 61.8 GB | A100 | 1 | $1.17/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-FP8 at its published (F8_E4M3) precision: 3× RTX 4090, at $0.441/hr per GPU ($1.32/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-FP8: common questions
Can GLM-4.5-Air-FP8 run on a single GPU?
No. At FP8 (native) it needs 123.5 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 4090, and it takes 3 of them.
Is GLM-4.5-Air-FP8 already quantized?
Yes. It is published in FP8, one byte per parameter, so the 123.5 GB figure is already a quantized footprint rather than a full-precision one. Only INT4 goes below it, at 61.8 GB. A GPU without FP8 tensor cores cannot run it as published, which is why cards here are matched on precision support and not on VRAM alone.
How many GPUs do I need to run GLM-4.5-Air-FP8?
3 at FP8 (native). It needs 123.5 GB of VRAM and the cheapest capable live offer is a 48.0 GB RTX 4090, so 3 of them come to $1.32/hr in total.
Does quantizing GLM-4.5-Air-FP8 lower the GPU bill?
Yes. At FP8 (native) the cheapest live fit is 3 RTX 4090 cards at $1.32/hr. At INT4 (quantized) it drops to one A100 at $1.17/hr, provided a quantized checkpoint exists for it.
Weight-to-VRAM math, the fit rules, and how live prices are normalized: full methodology.
More zai-org models
- GLM-OCR (1.3B, BF16)
- GLM-4.7-Flash (31.2B, BF16)
- GLM-5.2-FP8 (753.3B, F8_E4M3)
- GLM-5.2 (753.3B, BF16)
- GLM-5-FP8 (753.9B, F8_E4M3)
- GLM-5.3-Flash (321.3B, F8_E4M3)
Related reading: RTX 4090 pricing and specs, and The best GPUs for AI, ranked.
