MODEL · ASSET FILE
NVIDIA H100SXM nv.h100sxm
178 public observations · 54 real device-level IDs · rental range no quotes yet. Every record sits on an append-only public hash chain — independently replayable.
Observations
178
public records
Device IDs
54
real device-level identity (L0/L1/L2)
Model IDs
0
model-level identity (M)
Rental Range
no quotes yet
latest per platform
01
Rental quotes (latest per platform)
No quote observations for this model yet.
02
Observation mix
| Type | Count |
|---|---|
| listing | 178 |
03
Registered identities under this model
cvin:nv.h100sxm.sxm.80g:44ezj0fs5w0xknsx cvin:nv.h100sxm.sxm.80g:5eksh17ssno5i5uc cvin:nv.h100sxm.sxm.80g:5utyqo0p3ycw4rv6 cvin:nv.h100sxm.pcie.80g:2yise6c25nbno6vb cvin:nv.h100sxm.sxm.80g:4p4o5rpw5pklj9zi cvin:nv.h100sxm.sxm.80g:67bi2bly09ou1ahn cvin:nv.h100sxm.sxm.80g:3upnbpktz6mihwem cvin:nv.h100sxm.sxm.80g:lyviuvctli5a34jd… 54 total
04
Cost questions for this model
Hourly price is only half the decision. These pages work out what the card costs for a concrete job — month of rental, an inference workload, a fine-tune, a training run — using the same live prices shown on this page.
Which cloud provider is cheapest for an H100? What does an 8-GPU H100 server cost per month? How far apart can the same GPU be priced across providers? How many H100s and how much money to train a 70B model? A100 or H100: which is cheaper per unit of compute? What does Llama-3 70B inference cost per million tokens?