How much VRAM does GLM-4.5 355B (32B active) Chat need?
GLM-4.5 355B (32B active) Chat needs about 211.2 GB of VRAM at Q4_K_M, or 369.6 GB at Q8_0 and 695.1 GB at FP16. No single GPU in our set holds it even at Q4_K_M; it needs multi-GPU or offload. Weights plus runtime overhead. Budget separately for KV cache, which at 128K is a material share.
Weights-plus-runtime footprint across the six most common quantizations. KV cache is context-dependent and comes on top. At 128K it is gigabytes on its own, before the weights; size it precisely in the GPU Memory Calculator.
This is a mixture-of-experts model: all 355B parameters must sit in memory, but only about 32B are active per token — so it loads like a 355B model and runs closer to a 32B one.
Can your GPU run GLM-4.5 355B (32B active) Chat?
Pick your card — the answer below is computed by the same engine that produces every figure on this page.
H200 141GB cannot hold GLM-4.5 355B (32B active) Chat at Q4_K_M: it needs 211.2 GB against 133.9 GB usable; a heavier quantization, a larger card, or splitting across GPUs are the ways forward; as a mixture-of-experts model all 355B parameters must be resident even though only 32B are active per token.
Weights plus runtime overhead. KV cache is not included — it depends on context length and this model’s published architecture.
| Weights (Q4_K_M) | 200.4 GB |
| Runtime overhead | 10.8 GB |
| Estimated total | 211.2 GB |
| Usable VRAM (95% of 141 GB) | 133.9 GB |
Across every GPU we track
- RTX 3060 12GBNot enough-199.8 GB
- RTX 4060 Ti 16GBNot enough-196.0 GB
- RTX 4070 Ti SUPERNot enough-196.0 GB
- RTX 3090Not enough-188.4 GB
- RTX 4090Not enough-188.4 GB
- RTX 5090Not enough-180.8 GB
- A100 40GBNot enough-173.2 GB
- RTX A6000Not enough-165.6 GB
- L40SNot enough-165.6 GB
- A100 80GBNot enough-135.2 GB
- H100 80GBNot enough-135.2 GB
- H200 141GBNot enough-77.3 GB
Estimates, not guarantees: real usage moves with runtime, driver, batch size and context. Size a specific context window in the GPU Memory Calculator.
How was this number calculated? Every figure here comes from Bitpute’s documented calculation methodology — parameters, precision, quantization, runtime overhead and usable VRAM, each formula written out in full.
Memory by quantization
| Quant | Weights | + runtime overhead |
|---|---|---|
| FP16 | 661.2 GB | 695.1 GB |
| Q8_0 | 351.3 GB | 369.6 GB |
| Q6_K | 271.1 GB | 285.4 GB |
| Q5_K_M | 235.2 GB | 247.7 GB |
| Q4_K_M | 200.4 GB | 211.2 GB |
| Q4_0 | 186.0 GB | 196.0 GB |
Overhead = 0.75 GB + 5% of weights (CUDA context, buffers). Add KV cache on top. Weights dominate at this size, but a 128K window still costs gigabytes.
Which quantization should you actually run?
GLM-4.5 355B (32B active) Chat is a multi-GPU or datacenter model: even Q4_K_M needs about 211.2 GB, beyond any single card. Plan for several 80 GB GPUs or a hosted endpoint; FP16 (695.1 GB) is server-only. As a mixture-of-experts model it must hold all 355B in VRAM but activates only about 32B per token, so it runs much faster than its footprint suggests — memory is the limit here, not speed.
Single-GPU fit at Q4_K_M (211.2 GB + KV)
RTX 3060 12GBRTX 4060 Ti 16GBRTX 4070 Ti SUPERRTX 3090RTX 4090RTX 5090A100 40GBRTX A6000L40SA100 80GBH100 80GBH200 141GB
Common questions
How much VRAM does GLM-4.5 355B (32B active) Chat need?
GLM-4.5 355B (32B active) Chat needs about 211.2 GB of VRAM at Q4_K_M, 369.6 GB at Q8_0, or 695.1 GB at FP16. That is weights plus about 0.75 GB of runtime overhead and 5% of weight size; KV cache is additional and depends on context length.
What GPU can run GLM-4.5 355B (32B active) Chat?
No single GPU in our set fits it at Q4_K_M; it needs multiple GPUs or CPU offload. At FP16 no single GPU in our set is sufficient.
Can GLM-4.5 355B (32B active) Chat run on 24 GB?
Not at Q4_K_M: it needs about 211.2 GB, more than the ~22.8 GB usable on a 24 GB card. Use a larger card, multiple GPUs, or offload.