SVCI Public GPU Rate Index
103.8
Up 3.8 points from Issue 002. The move reflects coverage and composition, not broad matched-provider repricing.
Compute Opacity Index
28.5
Weighted public-price opacity rose as Tier 2 coverage fell from ten providers to nine.
Coverage Score
0.7150
Weighted share of the panel resolving to a comparable fixed public rate.
Companion Dollar Basket
$6.9333
Absolute USD per GPU-hour companion, up 1.99 percent from Issue 002.
The readings

Headline readings.

Issue 003 index values
IndexValueBasis
SVCI Public GPU Rate Index103.8Headline composite, base = Issue 001; weights 0.45 / 0.45 / 0.10
Compute Opacity Index28.5100 times (1 minus CompositeCoverage)
Hyperscaler Premium2.7352xTier 1 median over the pooled non-hyperscaler median
Neocloud Discount0.6344Tier 2 median relative to the Tier 1 median
Price Floor2.8200Lower-quartile headline-rate floor marker
Coverage Score0.7150Weighted panel coverage

The Confidence Score remains unpublished pending the per-row confidence audit. All other headline readings are calculated from the locked Issue 003 data layer.

Free reader tools

Get the next SVCI issue when it publishes.

Receive the benchmark instead of hunting for it. The public issue stays open; the subscription delivers the Monday Compute Price Brief, the market reading that explains it, and buyer tools as they launch.

Check a GPU quote

Tier medians

The three tiers.

Median on-demand rate by tier, USD per GPU-hour
TierWeightPanel slotsN in windowMedian USD
Tier 1, Hyperscalers0.454410.5306
Tier 2, GPU clouds and neoclouds0.451893.8500
Tier 3, Marketplaces and aggregators0.10524.6200

Panel slots are the locked twenty-seven provider composition across three tiers. N in window is the count of in-window observations that resolve to a comparable public rate. Medians are computed from the official capture.


Panel coverage

What the panel shows.

SVCI Issue 003 reviews all twenty-seven headline-panel providers. Fifteen published comparable fixed public H100 SXM 80GB rates that cleared the inclusion gate. Twelve were captured and excluded because the public surface was a floor, marketplace, aggregator, sales gate, signed-in-only rate, or incompatible product mode.

The headline index rose to 103.8 because Tier 2 public-price coverage fell from ten providers to nine. The provider that dropped from the comparable set was Voltage Park, whose launch-issue fixed rate of $1.99 no longer appeared as a public fixed price.

The index tracks the panel. The medians use only comparable public rates.
Panel status, Issue 003
StatusCount
Headline-eligible public rates15
Captured and excluded12
Pending0
Total headline panel27
Coverage by tier
TierResolvedCoverage
Tier 1, Hyperscalers4 / 4100.0%
Tier 2, GPU clouds9 / 1850.0%
Tier 3, Marketplaces2 / 540.0%
Weighted Coverage Score0.7150
Compute Opacity Index28.5

The Rule I-16 publication gate is met: coverage is non-zero in all three tiers. The confidence audit remains separate from the public rate inclusion decision.


Matched-panel sensitivity

What moved, and what did not.

Among the nine Tier 2 providers that published a comparable fixed rate in both Issue 001 and Issue 003, the median was unchanged at $3.85. Together AI repriced lower, but remained above the median. The canonical rise to 103.8 is therefore a coverage and composition event, not a broad matched-provider price increase.

Noncanonical matched-panel sensitivity
MetricIssue 001Issue 003
Matched Tier 2 median, N=9$3.8500$3.8500
Matched-panel SVCI100.0100.0
Why Voltage Park did not enter the Issue 003 median. Voltage Park remained a Tier 2 panel member and stayed in the coverage denominator. Its launch issue exposed exact fixed on-demand tiers. In the Issue 003 capture, the dedicated pricing cards read “Contact for pricing”; the homepage retained only “Starting at $1.99/hr.” The visible “$2.75/GPU/Hr” appears inside a product-interface illustration, not a live public rate field. Because the current public surface did not provide one reproducible fixed point rate, the observation was captured and disclosed but excluded from the median.

Tier 3 disclosure

Thin coverage.

Tier 3 remains a two-of-five fixed-rate observation. RunPod and Paperspace enter the tier median. Vast.ai, TensorDock, and Shadeform do not because dynamic marketplace medians, starting floors, and aggregator pass-through prices are not comparable fixed provider rates. The 0.10 weight remains locked.


Evidence and corrections

How to check this issue.

Every published rate traces to a recorded source and evidence artifact. The public evidence manifest identifies the provider folders, screenshots, source notes, and SHA-256 hashes retained in the sealed package. The canonical data layer, calculation report, capture store, and reconciliation ledger are published with the issue.

The full screenshot archive remains the sealed public evidence package of record. Its manifest is published here so every retained artifact can be identified and hash-checked. Corrections remain permanently available in the corrections record.


Appendix A

Headline panel status.

The headline panel contains twenty-seven provider slots. Fifteen comparable fixed public H100 SXM 80GB rates enter the medians. Twelve captured rows are disclosed as non-inclusions. No row remains pending.

Tier 1, Hyperscalers, 4 of 4 priced
IDProviderStatusMedianReasonNote
E001AWSPricedYesPublic on-demand p5.48xlarge, normalized per GPU by 8
E002AzurePricedYesPublic ND96isr H100 v5 on-demand rate, normalized per GPU by 8
E003GCPPricedYesa3-highgpu-8g on-demand rate, normalized per GPU by 8
E004OraclePricedYesBM.GPU.H100.8 public rate, normalized per GPU by 8
Tier 2, GPU clouds and neoclouds, 9 of 18 priced
IDProviderStatusMedianReasonNote
E005CoreWeavePricedYesHGX H100 on-demand, normalized per GPU by 8
E006CrusoePricedYesH100 80GB HGX on-demand per GPU
E007LambdaPricedYesH100 SXM on-demand per GPU; tax footnote resolved
E008NebiusPricedYesH100 on-demand GPU-hour rateRate remained above the Tier 2 median.
E009Together AIPricedYesGPU Clusters H100 on-demand, explicitly per GPUTogether AI fell from the launch issue but remained above the median.
E010HyperstackPricedYesH100 SXM on-demand public rate
E011VerdaPricedYesH100 SXM5 80GB pay as you goDataCrunch now operates under the Verda brand.
E012Voltage ParkCaptured, excludedNoStarting-from floor and contact pricing, no comparable fixed rate
E013FluidstackCaptured, excludedNoNo public fixed rate; pricing route returned not found
E014NscaleCaptured, excludedNoNo public fixed rate; reserve flow only
E015IRENCaptured, excludedNoH100 80GB SKU shown, price is contact sales
E016Applied DigitalCaptured, excludedNoInfrastructure and AI-factory product, no public GPU rental ratePanel-fit review is carried in the governance register.
E017GMI CloudCaptured, excludedNoDedicated pricing pages show from $2.00, a floor rather than a fixed rate
E018Cudo ComputeCaptured, excludedNoExact H100 SXM 80GB SKU shown, quote on request
E019SpheronCaptured, excludedNoLive marketplace floor and spot pricing, not a fixed provider rate
E020JarvislabsPricedYesH100 SXM 80GB 1x on-demand rateDaylight recapture confirmed the original $2.69 rate.
E021HyperbolicCaptured, excludedNoNo public comparable fixed rate; public, enterprise, and signed-in modes differ
E022Atlas CloudPricedYesH100 SXM 80GB fixed on-demand rate, per GPU
Tier 3, Marketplaces and aggregators, 2 of 5 priced
IDProviderStatusMedianReasonNote
E023RunPodPricedYesSecure Cloud fixed H100 SXM on-demand rateCurrent official page verified June 25; visual reference is from June 15.
E024Vast.aiCaptured, excludedNoDynamic marketplace median and range, excluded under F016
E025PaperspacePricedYesPaperspace H100 on-demand documentation rate; 8x cross-checkLegacy $2.24 commitment and DigitalOcean $3.39 product were excluded.
E026TensorDockCaptured, excludedNoStarting-from $2.25 floor, not a fixed comparable rate
E027ShadeformCaptured, excludedNoAggregator pass-through rates, no Shadeform-owned fixed rate
Status roll-up
CategoryCount
Priced, enter medians15
Captured and excluded12
Pending0
Total headline panel27

By the Sorso View Editorial. Full construction at sorsoview.com/methodology.


Issue archive

Previous issue.

Issue 002 is preserved at an SVCI level of 100.0. Issue 001 remains the base issue.

Read Issue 002 or browse the issue archive.