Can I Run GLM-5 / GLM-5.1 on 96 GB system RAM?
Written by Jakub Rusinowski · Last updated May 1, 2026
Technically yes, but not recommended
It loads, but it is not worth running — GLM-5.1 72B at Q6_K fits in 96 GB system RAM's 76.8 GB, yet the memory bandwidth limits it to ~1.1 tok/s (estimated), well below usable interactive speed.
Confidence: medium · Recommended quantization: Q6_K · Estimated speed: ~1.1 tok/s
96 GB system RAM — what it gives a model
| Usable memory for models | 76.8 GB |
| Memory bandwidth | 90 GB/s |
GLM-5 / GLM-5.1 on 96 GB system RAM: memory by quantization
| Quant | Memory needed | Fits 76.8 GB? | Max context | Est. speed | Download |
|---|
| F16 | 147.2 GB | ✗ No | — | — | 144 GB |
| Q8_0 | 79.7 GB | ✗ No | — | — | 76.5 GB |
| Q6_K | 62.3 GB | ✓ Yes | 32K | ~1.1 tok/s | 59 GB |
| Q5_K_M | 54.2 GB | ✓ Yes | 64K | ~1.3 tok/s | 51 GB |
| Q4_K_M | 46.7 GB | ✓ Yes | 64K | ~1.5 tok/s | 43.5 GB |
| Q3_K_M | 33.9 GB | ✓ Yes | 64K | ~2.1 tok/s | 30.7 GB |
| Q2_K | 26.9 GB | ✓ Yes | 64K | ~2.7 tok/s | 23.7 GB |
Which GLM-5 / GLM-5.1 sizes fit
| Variant | Needs at Q4_K_M | Fits? | Est. speed |
|---|
| GLM-5 744B | 455.2 GB | ✗ Too large | — |
| GLM-5.1 72B | 46.7 GB | ✓ Fits | ~1.5 tok/s |
| GLM-5 32B | 22 GB | ✓ Fits | ~3.3 tok/s |
| GLM-5 9B | 7.4 GB | ✓ Fits | ~10.9 tok/s |
What to watch out for
- At ~1.1 tok/s this loads but is too slow for interactive use — expect roughly 55 seconds per 60 tokens.
- 1 larger variant of GLM-5 / GLM-5.1 does not fit and would need CPU offload or different hardware.
- This model does not publish its full attention configuration, so the KV-cache share of these figures is inferred from its parameter count rather than computed exactly.
- These figures assume CPU-only inference. Any discrete GPU, even an 8 GB one, will be several times faster for models that fit in its VRAM.
Recommended setup
llama.cpp (CPU build) or Ollama — both run without a GPU
How these numbers are calculated
- Memory need = quantized weights + KV cache at 8,192 tokens (f16) + 0.8 GB runtime overhead.
- 76.8 GB of the 96 GB is treated as usable for model weights (80% — the rest is the OS and running applications).
- DDR5-5600 dual channel at 89.6 GB/s peak. CPU decode is assumed to sustain 35% of that peak, because CPU inference is not purely bandwidth-bound — it also spends real time in compute and thread synchronisation. This figure is an assumption, not a fitted constant: no CPU measurement is in the calibration set.
- CPU-only inference: no GPU is assumed. A GPU of any size will beat these figures substantially.
- Throughput is a memory-bandwidth roofline estimate, not a measurement. It is labelled "estimated" everywhere it appears.
- KV cache is inferred from the parameter count because this model does not publish its full attention configuration.
FAQ
Can I run GLM-5 / GLM-5.1 on 96 GB system RAM?
It loads, but it is not worth running — GLM-5.1 72B at Q6_K fits in 96 GB system RAM's 76.8 GB, yet the memory bandwidth limits it to ~1.1 tok/s (estimated), well below usable interactive speed.
Which quantization of GLM-5 / GLM-5.1 should I use on 96 GB system RAM?
Q6_K — it needs about 62.3 GB of the 76.8 GB available, downloads as roughly 59 GB, and runs at an estimated 1.1 tokens/sec with up to 32K of context.
What limits GLM-5 / GLM-5.1 on 96 GB system RAM?
Memory bandwidth. The model fits, but at 89.6 GB/s it can only be read fast enough for roughly 1.1 tokens/sec.
Which runtime should I use?
llama.cpp (CPU build) or Ollama — both run without a GPU
Other RAM Capacities
Other Models on 96 GB system RAM
GLM-5 / GLM-5.1 on GPUs
What This Model Is Good At
Model & Tools
← Can I Run It? | GLM-5 / GLM-5.1 model page | Check your hardware