Can GLM 4 32B 0414 run on the H100 SXM?
Yes. At 8K context the best fit is F16, using about 61.8 GB of the 76.0 GB usable on the H100 SXM. Estimated speed ceiling: ~51 tokens/sec.
Memory by quant and context
| Quant | 4K ctx | 8K ctx | 32K ctx | Max tok/s (est., 8K) |
|---|---|---|---|---|
| 2-bit (Q2_K) • | 12.4 GB Fits | 12.6 GB Fits | 14.0 GB Fits | 273 |
| 3-bit (Q3_K_M) • | 15.7 GB Fits | 16.0 GB Fits | 17.4 GB Fits | 211 |
| 4-bit (Q4_K_M) • | 19.3 GB Fits | 19.5 GB Fits | 20.9 GB Fits | 170 |
| 5-bit (Q5_K_M) • | 22.4 GB Fits | 22.7 GB Fits | 24.1 GB Fits | 145 |
| 6-bit (Q6_K) • | 25.8 GB Fits | 26.1 GB Fits | 27.5 GB Fits | 125 |
| 8-bit (Q8_0) • | 33.2 GB Fits | 33.4 GB Fits | 34.8 GB Fits | 97 |
| 16-bit (F16/BF16) | 61.6 GB Fits | 61.8 GB Fits | 63.3 GB Fits | 51 |
• real GGUF file size from axteam/GLM-4-32B-0414-GGUF. Speed is a bandwidth ceiling (3350 GB/s), not a benchmark. Methodology.
More
GLM 4 32B 0414 on every GPU and Mac · Every model that runs on the H100 SXM · Other models that fit 64 GB VRAM · Try other settings in the calculator
Data: Hugging Face Hub (zai-org/GLM-4-32B-0414), NVIDIA specs. Data as of 2026-11-17.