GLM-4.5 Air
110B parameters · 12B active · native context 131,072 · glm4_moe
Estimated VRAM requirement
Weight memory uses parameter count × bits per weight, plus a quantization overhead factor. KV cache uses standard GQA formulas. Values are calculated estimates, not measured allocations.
| Quantization | Estimated weights | 2K total | 4K total | 8K total | 16K total | 32K total | 64K total | 128K total |
|---|---|---|---|---|---|---|---|---|
| FP16 | 209.9 GB | 227.8 GB | 228.1 GB | 228.9 GB | 230.3 GB | 233.2 GB | 238.9 GB | 250.4 GB |
| Q8 | 118.1 GB | 128.6 GB | 129.0 GB | 129.7 GB | 131.1 GB | 134.0 GB | 139.8 GB | 151.3 GB |
| Q6 | 92.0 GB | 100.4 GB | 100.8 GB | 101.5 GB | 102.9 GB | 105.8 GB | 111.6 GB | 123.1 GB |
| Q5 | 77.8 GB | 85.1 GB | 85.5 GB | 86.2 GB | 87.7 GB | 90.5 GB | 96.3 GB | 107.8 GB |
| Q4 | 64.8 GB | 71.1 GB | 71.5 GB | 72.2 GB | 73.6 GB | 76.5 GB | 82.3 GB | 93.8 GB |
| Q3 | 51.8 GB | 57.0 GB | 57.4 GB | 58.1 GB | 59.5 GB | 62.4 GB | 68.2 GB | 79.7 GB |
Find a GPU for this model
Run at the quantization and context selected below. Defaults are Q4 and 8K when you do not change the controls. Results are full-VRAM fits only — not a speed ranking.
- Run at
- Q4 · 8K context
- Required VRAM
- 72.2 GB
No fresh matched Amazon price is currently available for a full-VRAM fit. See compatible GPUs · Workstations · Rent a server · Buy vs rent.
Lowest-cost full-VRAM option
Sorted by lowest current Amazon price among GPUs that fit entirely in VRAM. This is not a performance ranking.
No compatible GPU currently has a fresh Amazon price.
More VRAM headroom
Sorted by leftover usable VRAM after the calculated requirement. Headroom is not tokens/sec.
NVIDIA B200 192 GB
- GPU VRAM
- 192 GB advertised
- Required VRAM
- 72.2 GB
- Usable GPU VRAM
- 172.8 GB
- VRAM headroom
- 100.6 GB — leftover usable memory after the calculated requirement, not a speed ranking
- Compatible
- Yes — fits entirely in VRAM
- Product SKUs
- 0 mapped · 0 Amazon-matched
- Amazon
- Check availability
NVIDIA H200 141 GB
- GPU VRAM
- 141 GB advertised
- Required VRAM
- 72.2 GB
- Usable GPU VRAM
- 126.9 GB
- VRAM headroom
- 54.7 GB — leftover usable memory after the calculated requirement, not a speed ranking
- Compatible
- Yes — fits entirely in VRAM
- Product SKUs
- 0 mapped · 0 Amazon-matched
- Amazon
- Check availability
NVIDIA GH200 96 GB
- GPU VRAM
- 96 GB advertised
- Required VRAM
- 72.2 GB
- Usable GPU VRAM
- 86.4 GB
- VRAM headroom
- 14.2 GB — leftover usable memory after the calculated requirement, not a speed ranking
- Compatible
- Yes — fits entirely in VRAM
- Product SKUs
- 0 mapped · 0 Amazon-matched
- Amazon
- Check availability
Compatible GPUs
Status is a calculated memory fit against usable VRAM (advertised × 0.9). Open a GPU to see product SKUs and Amazon listings.
Sources
Model specifications: official repository zai-org/GLM-4.5-Air · revision main. Compatibility: RigForAI calculator v1.0.0 (estimated VRAM). GPU products: Icecat. Amazon is a separate affiliate match and is not used as a spec source.