Z.ai · 106B (12B active) · Mixture of Experts
Consumer-friendly GLM MoE — 12B active, strong agentic & tool use Check if your GPU or Mac can run GLM-4.5 Air locally — 59.2 GB min, 98.7 GB recommended.
2025-07128K context
Mixture of Experts
Total experts: 128
Active experts: 8
Active params: 12.0B
| Quant | Bits | VRAM | Quality | Status |
|---|---|---|---|---|
| Q2_K | 2 | 34.4 GB | low | — |
| Q3_K_M | 3 | 48 GB | moderate | — |
| Q4_K_M | 4 | 54.8 GB | good | — |
| Q5_K_M | 5 | 68.4 GB | good | — |
| Q6_K | 6 | 81.9 GB | excellent | — |
| Q8_0 | 8 | 109.1 GB | excellent | — |
| F16 | 16 | 217.7 GB | lossless | — |
Can I run GLM-4.5 Air locally?
- Can I run GLM-4.5 Air locally?
- GLM-4.5 Air needs about 59.2 GB of memory at a minimum and 98.7 GB recommended. Open this page to grade it against your GPU or Mac, then run it with runai, Ollama or LM Studio.
- How much VRAM does GLM-4.5 Air need?
- At Q4_K_M, GLM-4.5 Air uses about 54.8 GB of VRAM. Higher quants need more memory; lower quants fit tighter cards with a quality tradeoff.