Published GPU catalog prices were steady this week: none of the matched prices changed by more than 1%. Across the nine providers we track, the lowest published price held in each of the four GPU cohorts this Pulse covers: H100 SXM at $3.90 per GPU-hour (Crusoe), H200 at $4.29 (Crusoe), B200 at $6.69 (Lambda) and A100 SXM4 at $2.30 (Crusoe). Most of the 118 newly visible listings are AWS: H200 instances in Sydney, London and Ohio, and RTX PRO 6000 instances in Thailand.
The market index moved more than published prices did, but its endpoints do not describe the daily path. From October 26 to November 1, the Ornn Compute Price Index (OCPI), a third-party index of GPU compute prices from Ornn and ICE, fell for H100 SXM, B200 and A100 SXM4. It ended flat for H200, despite a 16% fall between October 27 and October 30. The readings are in the OCPI table below, and the daily path and the gap to published prices follow the charts.
Read the gap, not just the move. Because published prices held while the index moved, the gap between them is the more useful signal for buyers. Where a published price sits well above the index, ask for quotes matched to your region and term rather than relying on the list price. Where it sits well below, confirm that capacity is actually orderable at that price before planning around it.
Across the October 26 to November 2 GPU pricing comparison, none of the matched public on-demand GPU prices across the providers we track changed by more than 1%. We compared 13,788 of this week’s 13,918 GPU price listings directly with last week’s prices for the same provider, region and GPU configuration. The other 130 listings had no exact prior match: 118 were newly visible and 12 had appeared in an earlier catalog.
The AIForge Works Weekly Pricing Pulse is informational and is not a live quote, capacity signal, availability guarantee, workload-specific recommendation or personalized account analysis. For workload-specific options, use Cloud Advisor. To inspect neighboring SKUs, regions or terms, open the Multi-Cloud Pricing Analyzer. To compare your own GPU spend with dated market evidence, use GPU Audit. Verify current price, capacity, availability, orderability, terms, fees and taxes directly with the provider before acting.
Ornn OCPI context
OCPI is the Ornn Compute Price Index (Ornn x ICE), a third-party index that tracks the traded price of compute rather than quoted or listed prices, shown as market context. Ornn measures on-demand rentals only; reserved-capacity, long-term and forward contracts are outside the scope of the indices. These are settled daily readings through November 1, 2026. Ornn index source.
GPU family
October 26
November 1
6-day change
October 2
30-day change
H100 SXM
$3.30
$3.02
-8.5%
$2.72
+11.0%
H200
$6.21
$6.21
0.0%
$6.31
-1.6%
B200
$9.72
$9.44
-2.9%
$7.88
+19.8%
A100 SXM4
$1.08
$1.06
-1.9%
$0.89
+19.1%
All readings are US dollars per GPU-hour. Catalog and OCPI prices are displayed rounded to cents. Spreads and percentage changes are approximate and may differ slightly when recomputed from the displayed rounded values.
Lowest observed catalog price: The lowest comparable public on-demand price found in our November 2 catalog across the nine providers we track.
Ornn OCPI context
OCPI and Lowest Observed Catalog Prices
Saved public catalog snapshot · November 2, 2026
Exact GPU cohort comparison
H100 SXM
Public on-demand catalog prices for the same GPU group, normalized per GPU-hour. The lowest price may come from a different region or instance size than the offer you need.
OCPI-H100 SXM
Ornn x ICE - settled November 1, 2026
$3.02/ GPU / hr
Lowest observed catalog price
Crusoe - Global pricing · November 2 catalog
$3.90/ GPU / hr
Above OCPI
Catalog checked
Nov 2
All 9 provider files scanned
Providers with a comparable catalog price 8 / 9
Lowest observed catalog prices by providerNovember 2 catalog · lowest price first within each group
ProviderLowest catalog priceSpread vs OCPIInstance / SKURecorded region
Hyperscalers4/4 matched
AWSHyperscalers
$6.88/GPU/hr
+$3.86/ +127.8%
p5.48xlarge8 GPUs
US East (N. Virginia)
OCIHyperscalers
$10.00/GPU/hr
+$6.98/ +231.1%
B98415-us-ashburn-18 GPUs
us-ashburn-1
GCPHyperscalers
$11.06/GPU/hr
+$8.04/ +266.3%
gcp-machine-a3-highgpu-8g8 GPUs
us-west1
AzureHyperscalers
$12.29/GPU/hr
+$9.27/ +307.0%
Standard_ND96isr_H100_v58 GPUs
westus3
Neoclouds4/5 matched
CrusoeNeoclouds
$3.90/GPU/hr
+$0.88/ +29.1%
h100-80gb-sxm-ib.8x8 GPUs
Global pricing
LambdaNeoclouds
$3.99/GPU/hr
+$0.97/ +32.1%
gpu_8x_h100_sxm58 GPUs
Region not published
NebiusNeoclouds
$4.50/GPU/hr
+$1.48/ +49.0%
gpu-h100-sxm/1gpu-16vcpu-200gb1 GPU
eu-north1
CoreWeaveNeoclouds
$6.16/GPU/hr
+$3.14/ +103.8%
hgx-h1008 GPUs
global
VultrNeoclouds
N/A
No comparable price
No confirmed comparable price in this saved catalog.
N/A
Spread = catalog price minus the November 1 OCPI of $3.02/GPU-hour; percentage = spread divided by OCPI. Orange is above OCPI; green is below. Catalog and OCPI prices are displayed rounded to cents. Spreads and percentage changes are approximate and may differ slightly when recomputed from the displayed rounded values. This compares catalog prices with an index, not region-matched quotes.
Lowest observed catalog price: The lowest comparable public on-demand price found in our November 2 catalog across the nine providers we track.
Sources and limits: The check scanned 6,411,798 pricing records and found 471 comparable price listings across the four GPU groups. GPU specifications, source-file hashes and per-GPU arithmetic were checked. Prices exclude spot, reserved/committed and fractional-GPU offers. These are saved-catalog prices; they do not establish current capacity or orderability.
Exact GPU cohort comparison
H200
Public on-demand catalog prices for the same GPU group, normalized per GPU-hour. The lowest price may come from a different region or instance size than the offer you need.
OCPI-H200
Ornn x ICE - settled November 1, 2026
$6.21/ GPU / hr
Lowest observed catalog price
Crusoe - Global pricing · November 2 catalog
$4.29/ GPU / hr
Below OCPI
Catalog checked
Nov 2
All 9 provider files scanned
Providers with a comparable catalog price 7 / 9
Lowest observed catalog prices by providerNovember 2 catalog · lowest price first within each group
ProviderLowest catalog priceSpread vs OCPIInstance / SKURecorded region
Hyperscalers4/4 matched
AWSHyperscalers
$7.91/GPU/hr
+$1.70/ +27.4%
p5en.48xlarge8 GPUs
US East (N. Virginia)
OCIHyperscalers
$10.00/GPU/hr
+$3.79/ +61.0%
B110519-us-ashburn-18 GPUs
us-ashburn-1
AzureHyperscalers
$10.60/GPU/hr
+$4.39/ +70.7%
Standard_ND96isr_H200_v58 GPUs
westus2
GCPHyperscalers
$10.60/GPU/hr
+$4.39/ +70.7%
gcp-machine-a3-ultragpu-8g8 GPUs
us-central1
Neoclouds3/5 matched
CrusoeNeoclouds
$4.29/GPU/hr
-$1.92/ -30.9%
h200-141gb-sxm-ib.8x8 GPUs
Global pricing
NebiusNeoclouds
$5.40/GPU/hr
-$0.81/ -13.0%
gpu-h200-sxm/1gpu-16vcpu-200gb1 GPU
eu-north1
CoreWeaveNeoclouds
$6.31/GPU/hr
+$0.10/ +1.5%
hgx-h2008 GPUs
global
LambdaNeoclouds
N/A
No comparable price
No confirmed comparable price in this saved catalog.
N/A
VultrNeoclouds
N/A
No comparable price
No confirmed comparable price in this saved catalog.
N/A
Spread = catalog price minus the November 1 OCPI of $6.21/GPU-hour; percentage = spread divided by OCPI. Orange is above OCPI; green is below. Catalog and OCPI prices are displayed rounded to cents. Spreads and percentage changes are approximate and may differ slightly when recomputed from the displayed rounded values. This compares catalog prices with an index, not region-matched quotes.
Lowest observed catalog price: The lowest comparable public on-demand price found in our November 2 catalog across the nine providers we track.
Sources and limits: The check scanned 6,411,798 pricing records and found 471 comparable price listings across the four GPU groups. GPU specifications, source-file hashes and per-GPU arithmetic were checked. Prices exclude spot, reserved/committed and fractional-GPU offers. These are saved-catalog prices; they do not establish current capacity or orderability.
Exact GPU cohort comparison
B200
Public on-demand catalog prices for the same GPU group, normalized per GPU-hour. The lowest price may come from a different region or instance size than the offer you need.
OCPI-B200
Ornn x ICE - settled November 1, 2026
$9.44/ GPU / hr
Lowest observed catalog price
Lambda - Region not published · November 2 catalog
$6.69/ GPU / hr
Below OCPI
Catalog checked
Nov 2
All 9 provider files scanned
Providers with a comparable catalog price 5 / 9
Lowest observed catalog prices by providerNovember 2 catalog · lowest price first within each group
ProviderLowest catalog priceSpread vs OCPIInstance / SKURecorded region
Hyperscalers2/4 matched
OCIHyperscalers
$14.00/GPU/hr
+$4.56/ +48.3%
B110978-us-ashburn-18 GPUs
us-ashburn-1
AWSHyperscalers
$14.24/GPU/hr
+$4.80/ +50.9%
p6-b200.48xlarge8 GPUs
US East (N. Virginia)
AzureHyperscalers
N/A
No comparable price
No confirmed comparable price in this saved catalog.
N/A
GCPHyperscalers
N/A
No comparable price
No confirmed comparable price in this saved catalog.
N/A
Neoclouds3/5 matched
LambdaNeoclouds
$6.69/GPU/hr
-$2.75/ -29.1%
gpu_8x_b200_sxm68 GPUs
Region not published
NebiusNeoclouds
$8.50/GPU/hr
-$0.94/ -10.0%
gpu-b200-sxm/1gpu-20vcpu-224gb1 GPU
us-central1
CoreWeaveNeoclouds
$8.60/GPU/hr
-$0.84/ -8.9%
nvidia-b2008 GPUs
global
CrusoeNeoclouds
N/A
No comparable price
No confirmed comparable price in this saved catalog.
N/A
VultrNeoclouds
N/A
No comparable price
No confirmed comparable price in this saved catalog.
N/A
Spread = catalog price minus the November 1 OCPI of $9.44/GPU-hour; percentage = spread divided by OCPI. Orange is above OCPI; green is below. Catalog and OCPI prices are displayed rounded to cents. Spreads and percentage changes are approximate and may differ slightly when recomputed from the displayed rounded values. This compares catalog prices with an index, not region-matched quotes.
Lowest observed catalog price: The lowest comparable public on-demand price found in our November 2 catalog across the nine providers we track.
Sources and limits: The check scanned 6,411,798 pricing records and found 471 comparable price listings across the four GPU groups. GPU specifications, source-file hashes and per-GPU arithmetic were checked. Prices exclude spot, reserved/committed and fractional-GPU offers. These are saved-catalog prices; they do not establish current capacity or orderability.
Exact GPU cohort comparison
A100 SXM4
Public on-demand catalog prices for the same GPU group, normalized per GPU-hour. The lowest price may come from a different region or instance size than the offer you need.
OCPI-A100 SXM4
Ornn x ICE - settled November 1, 2026
$1.06/ GPU / hr
Lowest observed catalog price
Crusoe - Global pricing · November 2 catalog
$2.30/ GPU / hr
Above OCPI
Catalog checked
Nov 2
All 9 provider files scanned
Providers with a comparable catalog price 6 / 9
Lowest observed catalog prices by providerNovember 2 catalog · lowest price first within each group
ProviderLowest catalog priceSpread vs OCPIInstance / SKURecorded region
Hyperscalers4/4 matched
AWSHyperscalers
$3.43/GPU/hr
+$2.37/ +223.7%
p4de.24xlarge8 GPUs
US East (N. Virginia)
OCIHyperscalers
$4.00/GPU/hr
+$2.94/ +277.4%
B95907-us-ashburn-18 GPUs
us-ashburn-1
AzureHyperscalers
$4.10/GPU/hr
+$3.04/ +286.4%
Standard_ND96amsr_A100_v48 GPUs
westus3
GCPHyperscalers
$5.07/GPU/hr
+$4.01/ +378.2%
gcp-machine-a2-ultragpu-1g1 GPU
us-central1
Neoclouds2/5 matched
CrusoeNeoclouds
$2.30/GPU/hr
+$1.24/ +117.0%
a100-80gb-sxm-ib.8x8 GPUs
Global pricing
LambdaNeoclouds
$2.79/GPU/hr
+$1.73/ +163.2%
gpu_8x_a100_80gb_sxm48 GPUs
Virginia, USA
CoreWeaveNeoclouds
N/A
No comparable price
No confirmed comparable price in this saved catalog.
N/A
NebiusNeoclouds
N/A
No comparable price
No confirmed comparable price in this saved catalog.
N/A
VultrNeoclouds
N/A
No comparable price
No confirmed comparable price in this saved catalog.
N/A
Spread = catalog price minus the November 1 OCPI of $1.06/GPU-hour; percentage = spread divided by OCPI. Orange is above OCPI; green is below. Catalog and OCPI prices are displayed rounded to cents. Spreads and percentage changes are approximate and may differ slightly when recomputed from the displayed rounded values. This compares catalog prices with an index, not region-matched quotes.
Lowest observed catalog price: The lowest comparable public on-demand price found in our November 2 catalog across the nine providers we track.
Sources and limits: The check scanned 6,411,798 pricing records and found 471 comparable price listings across the four GPU groups. GPU specifications, source-file hashes and per-GPU arithmetic were checked. Prices exclude spot, reserved/committed and fractional-GPU offers. These are saved-catalog prices; they do not establish current capacity or orderability.
OCPI vs Lowest Observed Catalog Prices by Week
OCPI index prices alongside the lowest observed catalog price for each GPU. Historical points retain their dated comparison basis.
OCPI indexCatalog price
OCPI indexCatalog price
H100 SXM
OCPI$3.02Catalog$3.90(Crusoe)
$0.88 above OCPI
OCPI indexCatalog price
H200
OCPI$6.21Catalog$4.29(Crusoe)
$1.92 below OCPI
OCPI indexCatalog price
B200
OCPI$9.44Catalog$6.69(Lambda)
$2.75 below OCPI
OCPI indexCatalog price
A100 SXM4
OCPI$1.06Catalog$2.30(Crusoe)
$1.24 above OCPI
The November 1 point pairs that day’s OCPI settlement with the November 2 catalog minimum. Earlier points retain their published evidence.
The index's daily path, October 26 to November 1
H100 SXM: its October 26 and 27 readings were its highest in more than five weeks. It fell to $2.95 on October 28 and then held between $2.92 and $3.03.
H200: it was the most volatile, not the most stable. It fell 16%, from $6.56 on October 27 to $5.49 on October 30, before returning to $6.21.
B200 and A100 SXM4: B200 eased after peaking on October 28, and A100 SXM4 after holding its $1.10 high on October 28 and 29; both remain up about 20% from October 2.
Published prices versus the index
The index reading is dated November 1 and the catalog November 2, and they are not matched by region, so treat these gaps as directional.
H100 SXM: the cheapest published price ($3.90) sits about 29% above the index, up from about 18% on October 26.
A100 SXM4: the cheapest published price is more than twice the index.
H200 and B200: the gap runs the other way. Their cheapest published prices sit about 30% below the index.
Week at a glance
We compared 13,788 of this week’s 13,918 GPU price listings directly with last week’s prices for the same provider, region and GPU configuration. 130 listings did not have an exact prior match; the details below explain those differences. 0 matched prices changed by more than 1%.
Prior: 13,79313,788 matched + 5 no longer present = 13,793
Crusoe publishes one on-demand rate per GPU model, with no regional pricing. Crusoe prices are shown as global, and a Crusoe rate change is counted once.
Collection-method correction: For this issue, we restated the October 26 comparison baseline to 13,793 comparable GPU price listings, from the 13,806 reported in the October 26 issue. That issue used a one-time adjusted collection for CoreWeave and Nebius. This week both providers were collected the standard way, so we compare this week with the standard collection from the same October 26 catalog. The 13-listing difference is 9 CoreWeave listings (the same prices, recorded once rather than separately for North America and Europe) and 4 Nebius listings in regions our standard collection does not cover. This changes the comparison baseline only; it does not mean listings appeared or disappeared, or that prices changed. The October 26 issue remains as published.
Provider roll-up
Provider
Class
Newly visible
Restored
Newly absent
Price moves >1%
AWS
hyperscaler
117
0
0
0
GCP
hyperscaler
1
0
0
0
Lambda
neocloud
0
12
5
0
Totals
118
12
5
0
No reportable exact GPU catalog movement was found for Azure, OCI, CoreWeave, Crusoe, Nebius or Vultr.
Newly visible listings
AWS
117 newly visible price listings in 9 groups
p5e.48xlarge in Asia Pacific (Sydney). H200 141 GB. 5 price variants. Public on-demand: from $82.28/instance-hour ($10.29/GPU-hour; OS: Linux; tenancy: Shared) to $86.14/instance-hour ($10.77/GPU-hour; OS: Red Hat Enterprise Linux with HA; tenancy: Shared).
g7e.12xlarge in Asia Pacific (Thailand). NVIDIA RTX PRO 6000 BSE 96 GB. 17 price variants. Public on-demand: from $14.20/instance-hour ($7.10/GPU-hour; OS: Linux; tenancy: Shared) to $34.40/instance-hour ($17.20/GPU-hour; OS: Windows; tenancy: Shared; preinstalled software: SQL Ent).
g7e.24xlarge in Asia Pacific (Thailand). NVIDIA RTX PRO 6000 BSE 96 GB. 17 price variants. Public on-demand: from $28.39/instance-hour ($7.10/GPU-hour; OS: Linux; tenancy: Shared) to $68.81/instance-hour ($17.20/GPU-hour; OS: Windows; tenancy: Shared; preinstalled software: SQL Ent).
g7e.2xlarge in Asia Pacific (Thailand). NVIDIA RTX PRO 6000 BSE 96 GB. 17 price variants. Public on-demand: from $5.76/instance-hour ($5.76/GPU-hour; OS: Linux; tenancy: Shared) to $9.13/instance-hour ($9.13/GPU-hour; OS: Windows; tenancy: Shared; preinstalled software: SQL Ent).
g7e.48xlarge in Asia Pacific (Thailand). NVIDIA RTX PRO 6000 BSE 96 GB. 17 price variants. Public on-demand: from $56.78/instance-hour ($7.10/GPU-hour; OS: Linux; tenancy: Shared) to $137.61/instance-hour ($17.20/GPU-hour; OS: Windows; tenancy: Shared; preinstalled software: SQL Ent).
g7e.4xlarge in Asia Pacific (Thailand). NVIDIA RTX PRO 6000 BSE 96 GB. 17 price variants. Public on-demand: from $6.85/instance-hour ($6.85/GPU-hour; OS: Linux; tenancy: Shared) to $13.59/instance-hour ($13.59/GPU-hour; OS: Windows; tenancy: Shared; preinstalled software: SQL Ent).
g7e.8xlarge in Asia Pacific (Thailand). NVIDIA RTX PRO 6000 BSE 96 GB. 17 price variants. Public on-demand: from $9.03/instance-hour ($9.03/GPU-hour; OS: Linux; tenancy: Shared) to $22.50/instance-hour ($22.50/GPU-hour; OS: Windows; tenancy: Shared; preinstalled software: SQL Ent).
p5e.48xlarge in EU (London). H200 141 GB. 5 price variants. Public on-demand: from $82.28/instance-hour ($10.29/GPU-hour; OS: Linux; tenancy: Shared) to $86.14/instance-hour ($10.77/GPU-hour; OS: Red Hat Enterprise Linux with HA; tenancy: Shared).
p5e.48xlarge in US East (Ohio). H200 141 GB. 5 price variants. Public on-demand: from $63.30/instance-hour ($7.91/GPU-hour; OS: Linux; tenancy: Shared) to $67.16/instance-hour ($8.39/GPU-hour; OS: Red Hat Enterprise Linux with HA; tenancy: Shared).
GCP
1 newly visible price listings
C0D3-1B32-1712 in southamerica-east1. NVIDIA RTX PRO 6000 BSE 96 GB. Public on-demand GPU component price: $1.57/GPU-hour. vCPU, memory and other charges are billed separately, so this is not a whole-instance price.
Catalog changes in detail
Restored or expanded regional observations
Lambda: gpu_8x_a100 in Arizona, USA at $1.99 per GPU-hour.
Lambda: gpu_8x_a100 in California, USA at $1.99 per GPU-hour.
Lambda: gpu_1x_h100_sxm5 in Central Texas, USA at $4.29 per GPU-hour.
Lambda: gpu_2x_h100_sxm5 in Central Texas, USA at $4.19 per GPU-hour.
Lambda: gpu_8x_a100 in Germany at $1.99 per GPU-hour.
Lambda: gpu_8x_a100_80gb_sxm4 in Illinois, USA at $2.79 per GPU-hour.
Lambda: gpu_8x_a100 in Israel at $1.99 per GPU-hour.
Lambda: gpu_1x_a6000 in North Texas, USA at $1.09 per GPU-hour.
Lambda: gpu_2x_h100_sxm5 in North Texas, USA at $4.19 per GPU-hour.
Lambda: gpu_4x_h100_sxm5 in North Texas, USA at $4.09 per GPU-hour.
Lambda: gpu_8x_a100 in Osaka, Japan at $1.99 per GPU-hour.
Lambda: gpu_8x_a100_80gb_sxm4 in Virginia, USA at $2.79 per GPU-hour.
Prior regional observations not seen this week
Lambda: gpu_1x_a6000 in Region not published at $1.09 per GPU-hour.
Lambda: gpu_2x_h100_sxm5 in Region not published at $4.19 per GPU-hour.
Lambda: gpu_4x_h100_sxm5 in Region not published at $4.09 per GPU-hour.
Lambda: gpu_8x_a100 in Region not published at $1.99 per GPU-hour.
Lambda: gpu_8x_a100_80gb_sxm4 in Region not published at $2.79 per GPU-hour.
Same SKUs under different region labels
The following SKUs appear in both regional observation lists at the same recorded per-GPU price, GPU label and GPU count. These are separate region-qualified observations, not evidence that the underlying SKUs disappeared. The records alone do not establish whether the locations changed or their labels changed; unknown regions remain unknown. The counts retain the original exact comparison keys.
Lambda: gpu_1x_a6000; prior labels: Region not published; current labels: North Texas, USA.
Lambda: gpu_2x_h100_sxm5; prior labels: Region not published; current labels: Central Texas, USA, North Texas, USA.
Lambda: gpu_4x_h100_sxm5; prior labels: Region not published; current labels: North Texas, USA.
Lambda: gpu_8x_a100; prior labels: Region not published; current labels: Arizona, USA, California, USA, Germany, Israel, Osaka, Japan.
Lambda: gpu_8x_a100_80gb_sxm4; prior labels: Region not published; current labels: Illinois, USA, Virginia, USA.
Catalog visibility does not establish a product launch, retirement, or available capacity.
Evidence and methodology
We compare public on-demand prices for the same provider, region and GPU configuration. Prices are normalized per GPU-hour and rounded to cents. The lowest observed comparable catalog price excludes spot, reserved or committed terms, software-only charges and fractional GPUs. Catalog prices are saved evidence, not live quotes or guarantees of orderability.
Ornn has since revised its published OCPI history. Its October 26 readings changed from $3.17 to $3.30 for H100 SXM, from $9.76 to $9.72 for B200 and from $1.09 to $1.08 for A100 SXM4. This issue compares both weeks on Ornn’s current history, and the October 26 point in the OCPI chart uses the revised reading. Earlier chart points are shown as published. The October 26 issue remains as published.