← All GPUs / GPU detail
Live pricing intelligence

H100 SXM price by cloud provider

Track H100 SXM price across Vast.ai, AWS, Azure, and more. The lowest tracked H100 SXM median right now is $2.20/hr on Vast.ai (on-demand).

80GB HBM3 Hopper Flagship tier Provider price tracking
Lowest Median
Providers
Median Range
across recent provider rows
Data Points

What does H100 SXM cost in the cloud?

H100 SXM rental price changes with provider supply, region, and pricing model. This page gives you one place to compare recent H100 SXM rates across 7 providers; older snapshots remain visible with freshness context.

Use the latest pricing table for the newest market snapshot, then read the chart to see whether today's rate looks like a short-term discount or part of a broader trend.

Cheapest provider right now

H100 SXM lowest tracked median summary

The lowest tracked H100 SXM median right now is $2.20/hr on Vast.ai (on-demand). Current tracked median rates run up to $12.29/hr across 7 recent providers.

Methodology and freshness

How we normalize price and freshness

getflops.ai converts each provider's latest listing into a per-GPU hourly rate, keeps the newest snapshot per pricing type, and charts historical medians so you can separate temporary discounts from durable market moves.

H100 SXM price FAQ

What is the lowest tracked H100 SXM median right now?

The lowest tracked H100 SXM median right now is $2.20/hr on Vast.ai (on-demand). Current tracked median rates run up to $12.29/hr across 7 recent providers.

Why does H100 SXM rental price vary by provider?

H100 SXM pricing moves with supply, reservation model, spot availability, included networking or storage, and how much competition each provider has for the same GPU inventory. A lower hourly price can also reflect community or spot capacity instead of fully reserved on-demand instances.

Which providers are tracked for H100 SXM price comparisons?

We currently track 7 providers for H100 SXM: Vast.ai, AWS, Azure, GCP, Lambda, RunPod, and Oracle.

How fresh is the data on this H100 SXM price page?

Each provider is normalized into per-GPU hourly pricing and stored as its latest snapshot. The freshest H100 SXM row on this page is from Jul 31, 2026, and collectors run on a daily cadence.

H100 SXM prices by workload

Jump from the full GPU market into workload-specific provider rankings for inference, batch jobs, fine-tuning, and training.

Market median with provider overlays

Start with the blended market line, then overlay specific providers to see spreads, discounting, and outliers over time.

Latest H100 SXM price by provider

Recent rows rank first by median price; older snapshots stay visible with freshness context.

Loading current market data…
Provider Type Min Median Max Offers Last Updated
Loading...

Current marketplace listings

Raw Vast.ai offers are shown with host reliability, location, network, and the quality filter used for aggregate snapshots.

Loading Vast.ai offers…
Price Host quality GPUs Location CPU / RAM Network Offer
Loading...

Watch any H100 SXM price

Get notified when a tracked provider lists H100 SXM at or below your target hourly price. The editable target starts 10% below the lowest recent on-demand median.

Browse every H100 SXM price comparison

Open a side-by-side view to compare specs, price trends, and provider coverage.

Hand this runbook to Claude Code or Codex

Open a terminal in the repository where you want the deployment files, start claude or codex, then paste this prompt. It asks the agent to verify sources and stop before it creates billable infrastructure.

Deployment prompt
Use the infrastructure or model context on this page to create a reproducible open-model deployment. Use this guide as the starting context: https://www.getflops.ai/gpu/H100%20SXM. Read the linked model card and provider documentation before choosing hardware or runtime settings. Open every linked primary source and flag any mismatch instead of guessing. Create a deployment folder containing README.md, .env.example with no secrets, a pinned start script or infrastructure manifest, and smoke-test.sh. Make the endpoint OpenAI-compatible where the runtime supports it. Run local/static validation, estimate the billable resources, and stop before provisioning paid infrastructure until I approve.
Guardrails included No secrets in files · verify primary docs · approval before spend