RunsOnMy ModelsGPUs & MacsBy VRAMAPI pricesVRAM calculatorNew

Can Nemotron 3.5 Parse run on the H100 SXM?

Yes. At 8K context the best fit is F16, using about 1.9 GB of the 76.0 GB usable on the H100 SXM. Estimated speed ceiling: ~2563 tokens/sec.

Memory by quant and context

Quant4K ctx8K ctxMax tok/s (est., 8K)
2-bit (Q2_K)0.8 GB Fits0.9 GB Fits14162
3-bit (Q3_K_M)0.9 GB Fits1.0 GB Fits10866
4-bit (Q4_K_M)1.0 GB Fits1.1 GB Fits8688
5-bit (Q5_K_M)1.0 GB Fits1.1 GB Fits7454
6-bit (Q6_K)1.1 GB Fits1.2 GB Fits6476
8-bit (Q8_0) •1.4 GB Fits1.5 GB Fits3867
16-bit (F16/BF16) •1.8 GB Fits1.9 GB Fits2563

• real GGUF file size from SkyContact/Nemotron-3.5-Parse-GGUF. Speed is a bandwidth ceiling (3350 GB/s), not a benchmark. Methodology.

More

Nemotron 3.5 Parse 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 (nvidia/Nemotron-3.5-Parse), NVIDIA specs. Data as of 2026-11-17.