H100 SXM
NVIDIA · Data Center GPU · 2023
Performance Benchmarks
Image Generation all sourced rows →
| Model | Metric | Value | Source |
|---|---|---|---|
| SDXL Turbo FP16 | images per minute |
200
img/min
ComfyUI · 2025-06-01
|
Manufacturer benchmark NVIDIA Official Specs ↗ |
LLM Inference all sourced rows →
| Model | Metric | Value | Source |
|---|---|---|---|
|
Llama-3-70B
Q4 80GB HBM3 — editorial estimate pending article-level source verification |
tokens per second |
90
tok/s
est llama.cpp · 2025-06-01
|
Estimated MLPerf Inference Results ↗ |
|
Llama-3-8B
Q4 80GB HBM3 — editorial estimate pending article-level source verification |
tokens per second |
500
tok/s
est llama.cpp · 2025-06-01
|
Estimated NVIDIA Official Specs ↗ |
Full Specifications
Related GPUs
Common Questions
Can the H100 SXM run local LLMs?
Yes, if the model fits in its 80 GB of memory — VRAM capacity decides which models load at all, while its 3350 GB/s of bandwidth decides how fast tokens generate. With 24 GB or more it can hold large quantized models such as 70B-class LLMs in reduced precision, and our model pages list exact VRAM fits per model.
How much VRAM does the H100 SXM have?
H100 SXM has 80 GB of video memory with 3350 GB/s of memory bandwidth and a 700 W TDP , per the specification values in our hardware database.
Is the H100 SXM worth it for AI workloads?
Its value depends on whether your models fit in 80 GB. Compare its VRAM and bandwidth against your target models with our VRAM calculator, or see how it stacks up in head-to-head comparisons.
Guides & Tools
CompareAIHardware.com participates in the Amazon Associates program. As an Amazon Associate, we earn from qualifying purchases.