Ranked using workload compatibility, placement, observed pricing and source evidence. Unknown requirements remain open for provider confirmation; results are not reservations.
Placement & cost estimate
16 of 16 GPUs in the planning allocation · $70.56/hour.
Recorded capacity covers 0 GPUs; 16 use configuration limits, not confirmed stock. Remaining shortfall: 0.
16 allocated GPUs rely on configuration evidence rather than confirmed aggregate live capacity.
Best match
DigitalOcean · H100 SXM
New York Metro
Available in documented datacenter
$4.41 USD / GPU-hourTotal not estimated: add the GPU count and duration · Grade A
1 requirements need confirmation. Compute-only estimate; additional charges and final terms may apply.
Why this match · economics & evidence
GPU requirement matches the tracked accelerator.
Request exceeds the per-configuration maximum of 1 GPUs; multiple configurations would be required and aggregate capacity is not confirmed.
Billed planning month is 730 hours unless the workload uses an explicit run duration. Utilization assumption is 70.0%. No egress volume was requested, so egress cost is excluded from the scenario. This scenario models compute commitments, requested egress, storage, setup, interconnect and entered overhead; taxes and unlisted charges require contract review. Minimum term is undisclosed; no complete contractual total is claimed. Billed run duration is not specified or derivable. Setup fee is undisclosed or invalid; it is not assumed to be zero. No applicable public comparable performance benchmark or explicit throughput assumption is available.
Opens the provider's own site. Reconfirm price and availability there before you commit.
Offering brief · take this to the provider
A sourced summary of this catalogue record: original price and unit, normalised price, confidence grade and check date. It is not a quote or a reservation.
# Kovara offering brief
# Generated 2026-11-17T14:33:26.998Z. Values are catalogue records, not a reservation or a quote.
# Reconfirm price and availability with the provider before you commit.
kovara:
offering_id: "OFR-0647"
provider:
id: "PRV0013"
name: "DigitalOcean"
url: "https://www.digitalocean.com/pricing/gpu-droplets"
gpu:
id: "GPU0001"
name: "NVIDIA H100 SXM"
vram_gb: 80
region:
id: "REG0007"
name: "New York Metro"
country: "United States"
requested_gpus: 16
terms:
deployment_type: "GPU Droplet"
contract_type: "On-Demand"
billing_model: "Per second"
minimum_term_months: null
price:
original:
value: 4.41
currency: "USD"
unit: "instance-hour"
normalized_usd_per_gpu_hour: 4.41
availability:
status: "Available in documented datacenter"
recorded_gpu_count: null
evidence:
confidence_grade: "A"
last_checked: "2026-11-17T12:22:09.87+00:00"
Match 2
DigitalOcean · H100 SXM
New York Metro
Available in documented datacenter
$4.41 USD / GPU-hourTotal not estimated: add the GPU count and duration · Grade A
1 requirements need confirmation. Compute-only estimate; additional charges and final terms may apply.
Why this match · economics & evidence
GPU requirement matches the tracked accelerator.
Request exceeds the per-configuration maximum of 8 GPUs; multiple configurations would be required and aggregate capacity is not confirmed.
Billed planning month is 730 hours unless the workload uses an explicit run duration. Utilization assumption is 70.0%. No egress volume was requested, so egress cost is excluded from the scenario. This scenario models compute commitments, requested egress, storage, setup, interconnect and entered overhead; taxes and unlisted charges require contract review. Minimum term is undisclosed; no complete contractual total is claimed. Billed run duration is not specified or derivable. Setup fee is undisclosed or invalid; it is not assumed to be zero. No applicable public comparable performance benchmark or explicit throughput assumption is available.
Opens the provider's own site. Reconfirm price and availability there before you commit.
Offering brief · take this to the provider
A sourced summary of this catalogue record: original price and unit, normalised price, confidence grade and check date. It is not a quote or a reservation.
# Kovara offering brief
# Generated 2026-11-17T14:33:26.998Z. Values are catalogue records, not a reservation or a quote.
# Reconfirm price and availability with the provider before you commit.
kovara:
offering_id: "OFR-0648"
provider:
id: "PRV0013"
name: "DigitalOcean"
url: "https://www.digitalocean.com/pricing/gpu-droplets"
gpu:
id: "GPU0001"
name: "NVIDIA H100 SXM"
vram_gb: 80
region:
id: "REG0007"
name: "New York Metro"
country: "United States"
requested_gpus: 16
terms:
deployment_type: "GPU Droplet"
contract_type: "On-Demand"
billing_model: "Per second"
minimum_term_months: null
price:
original:
value: 35.28
currency: "USD"
unit: "instance-hour"
normalized_usd_per_gpu_hour: 4.41
availability:
status: "Available in documented datacenter"
recorded_gpu_count: null
evidence:
confidence_grade: "A"
last_checked: "2026-11-17T12:22:09.87+00:00"
Match 3
DigitalOcean · H100 SXM
Toronto
Available in documented datacenter
$4.41 USD / GPU-hourTotal not estimated: add the GPU count and duration · Grade A
1 requirements need confirmation. Compute-only estimate; additional charges and final terms may apply.
Why this match · economics & evidence
GPU requirement matches the tracked accelerator.
Request exceeds the per-configuration maximum of 1 GPUs; multiple configurations would be required and aggregate capacity is not confirmed.
Billed planning month is 730 hours unless the workload uses an explicit run duration. Utilization assumption is 70.0%. No egress volume was requested, so egress cost is excluded from the scenario. This scenario models compute commitments, requested egress, storage, setup, interconnect and entered overhead; taxes and unlisted charges require contract review. Minimum term is undisclosed; no complete contractual total is claimed. Billed run duration is not specified or derivable. Setup fee is undisclosed or invalid; it is not assumed to be zero. No applicable public comparable performance benchmark or explicit throughput assumption is available.
Opens the provider's own site. Reconfirm price and availability there before you commit.
Offering brief · take this to the provider
A sourced summary of this catalogue record: original price and unit, normalised price, confidence grade and check date. It is not a quote or a reservation.
# Kovara offering brief
# Generated 2026-11-17T14:33:26.998Z. Values are catalogue records, not a reservation or a quote.
# Reconfirm price and availability with the provider before you commit.
kovara:
offering_id: "OFR-0651"
provider:
id: "PRV0013"
name: "DigitalOcean"
url: "https://www.digitalocean.com/pricing/gpu-droplets"
gpu:
id: "GPU0001"
name: "NVIDIA H100 SXM"
vram_gb: 80
region:
id: "REG0014"
name: "Toronto"
country: "Canada"
requested_gpus: 16
terms:
deployment_type: "GPU Droplet"
contract_type: "On-Demand"
billing_model: "Per second"
minimum_term_months: null
price:
original:
value: 4.41
currency: "USD"
unit: "instance-hour"
normalized_usd_per_gpu_hour: 4.41
availability:
status: "Available in documented datacenter"
recorded_gpu_count: null
evidence:
confidence_grade: "A"
last_checked: "2026-11-17T12:23:09.684+00:00"
Match 4
DigitalOcean · H100 SXM
Toronto
Available in documented datacenter
$4.41 USD / GPU-hourTotal not estimated: add the GPU count and duration · Grade A
1 requirements need confirmation. Compute-only estimate; additional charges and final terms may apply.
Why this match · economics & evidence
GPU requirement matches the tracked accelerator.
Request exceeds the per-configuration maximum of 8 GPUs; multiple configurations would be required and aggregate capacity is not confirmed.
Billed planning month is 730 hours unless the workload uses an explicit run duration. Utilization assumption is 70.0%. No egress volume was requested, so egress cost is excluded from the scenario. This scenario models compute commitments, requested egress, storage, setup, interconnect and entered overhead; taxes and unlisted charges require contract review. Minimum term is undisclosed; no complete contractual total is claimed. Billed run duration is not specified or derivable. Setup fee is undisclosed or invalid; it is not assumed to be zero. No applicable public comparable performance benchmark or explicit throughput assumption is available.
Opens the provider's own site. Reconfirm price and availability there before you commit.
Offering brief · take this to the provider
A sourced summary of this catalogue record: original price and unit, normalised price, confidence grade and check date. It is not a quote or a reservation.
# Kovara offering brief
# Generated 2026-11-17T14:33:26.998Z. Values are catalogue records, not a reservation or a quote.
# Reconfirm price and availability with the provider before you commit.
kovara:
offering_id: "OFR-0652"
provider:
id: "PRV0013"
name: "DigitalOcean"
url: "https://www.digitalocean.com/pricing/gpu-droplets"
gpu:
id: "GPU0001"
name: "NVIDIA H100 SXM"
vram_gb: 80
region:
id: "REG0014"
name: "Toronto"
country: "Canada"
requested_gpus: 16
terms:
deployment_type: "GPU Droplet"
contract_type: "On-Demand"
billing_model: "Per second"
minimum_term_months: null
price:
original:
value: 35.28
currency: "USD"
unit: "instance-hour"
normalized_usd_per_gpu_hour: 4.41
availability:
status: "Available in documented datacenter"
recorded_gpu_count: null
evidence:
confidence_grade: "A"
last_checked: "2026-11-17T12:23:09.684+00:00"
Match 5
Google Cloud · H100 SXM
Iowa
Public price; quota/capacity varies
$11.06 USD / GPU-hourTotal not estimated: add the GPU count and duration · Grade A
1 requirements need confirmation. Compute-only estimate; additional charges and final terms may apply.
Why this match · economics & evidence
GPU requirement matches the tracked accelerator.
Request exceeds the per-configuration maximum of 8 GPUs; multiple configurations would be required and aggregate capacity is not confirmed.
Billed planning month is 730 hours unless the workload uses an explicit run duration. Utilization assumption is 70.0%. No egress volume was requested, so egress cost is excluded from the scenario. This scenario models compute commitments, requested egress, storage, setup, interconnect and entered overhead; taxes and unlisted charges require contract review. Minimum term is undisclosed; no complete contractual total is claimed. Billed run duration is not specified or derivable. No applicable public comparable performance benchmark or explicit throughput assumption is available.
Opens the provider's own site. Reconfirm price and availability there before you commit.
Offering brief · take this to the provider
A sourced summary of this catalogue record: original price and unit, normalised price, confidence grade and check date. It is not a quote or a reservation.
# Kovara offering brief
# Generated 2026-11-17T14:33:26.998Z. Values are catalogue records, not a reservation or a quote.
# Reconfirm price and availability with the provider before you commit.
kovara:
offering_id: "OFF000015"
provider:
id: "PRV0007"
name: "Google Cloud"
url: "https://cloud.google.com/products/compute/pricing/accelerator-optimized"
gpu:
id: "GPU0001"
name: "NVIDIA H100 SXM"
vram_gb: 80
region:
id: "REG0055"
name: "Iowa"
country: "United States"
requested_gpus: 16
terms:
deployment_type: "Virtual Machine"
contract_type: "On-Demand"
billing_model: "Hourly"
minimum_term_months: null
price:
original:
value: 88.490000119
currency: "USD"
unit: "instance-hour"
normalized_usd_per_gpu_hour: 11.061250014875
availability:
status: "Public price; quota/capacity varies"
recorded_gpu_count: null
evidence:
confidence_grade: "A"
last_checked: "2026-11-17T10:12:10.05+00:00"
Match 6
Google Cloud · H100 SXM
Iowa
Public price; quota/capacity varies
$11.68 USD / GPU-hourTotal not estimated: add the GPU count and duration · Grade A
1 requirements need confirmation. Compute-only estimate; additional charges and final terms may apply.
Why this match · economics & evidence
GPU requirement matches the tracked accelerator.
Request exceeds the per-configuration maximum of 8 GPUs; multiple configurations would be required and aggregate capacity is not confirmed.
Billed planning month is 730 hours unless the workload uses an explicit run duration. Utilization assumption is 70.0%. No egress volume was requested, so egress cost is excluded from the scenario. This scenario models compute commitments, requested egress, storage, setup, interconnect and entered overhead; taxes and unlisted charges require contract review. Minimum term is undisclosed; no complete contractual total is claimed. Billed run duration is not specified or derivable. No applicable public comparable performance benchmark or explicit throughput assumption is available.
Opens the provider's own site. Reconfirm price and availability there before you commit.
Offering brief · take this to the provider
A sourced summary of this catalogue record: original price and unit, normalised price, confidence grade and check date. It is not a quote or a reservation.
# Kovara offering brief
# Generated 2026-11-17T14:33:26.998Z. Values are catalogue records, not a reservation or a quote.
# Reconfirm price and availability with the provider before you commit.
kovara:
offering_id: "OFF000014"
provider:
id: "PRV0007"
name: "Google Cloud"
url: "https://cloud.google.com/products/compute/pricing/accelerator-optimized"
gpu:
id: "GPU0001"
name: "NVIDIA H100 SXM"
vram_gb: 80
region:
id: "REG0055"
name: "Iowa"
country: "United States"
requested_gpus: 16
terms:
deployment_type: "Virtual Machine"
contract_type: "On-Demand"
billing_model: "Hourly"
minimum_term_months: null
price:
original:
value: 93.400712807
currency: "USD"
unit: "instance-hour"
normalized_usd_per_gpu_hour: 11.675089100875
availability:
status: "Public price; quota/capacity varies"
recorded_gpu_count: null
evidence:
confidence_grade: "A"
last_checked: "2026-11-17T10:12:10.05+00:00"
Match 7
Microsoft Azure · H100 NVL 94GB
Virginia
Unconfirmed
$6.98 USD / GPU-hourTotal not estimated: add the GPU count and duration · Grade B
1 requirements need confirmation. Compute-only estimate; additional charges and final terms may apply.
Why this match · economics & evidence
GPU requirement matches the tracked accelerator.
Quantity compatibility cannot be confirmed from current capacity evidence.
Billed planning month is 730 hours unless the workload uses an explicit run duration. Utilization assumption is 70.0%. No egress volume was requested, so egress cost is excluded from the scenario. This scenario models compute commitments, requested egress, storage, setup, interconnect and entered overhead; taxes and unlisted charges require contract review. Minimum term is undisclosed; no complete contractual total is claimed. Billed run duration is not specified or derivable. Setup fee is undisclosed or invalid; it is not assumed to be zero. No applicable public comparable performance benchmark or explicit throughput assumption is available.
Opens the provider's own site. Reconfirm price and availability there before you commit.
Offering brief · take this to the provider
A sourced summary of this catalogue record: original price and unit, normalised price, confidence grade and check date. It is not a quote or a reservation.
# Kovara offering brief
# Generated 2026-11-17T14:33:26.998Z. Values are catalogue records, not a reservation or a quote.
# Reconfirm price and availability with the provider before you commit.
kovara:
offering_id: "OFF000145"
provider:
id: "PRV0001"
name: "Microsoft Azure"
url: "https://azure.microsoft.com/en-us/pricing/details/virtual-machines/linux/"
gpu:
id: "GPU0002"
name: "NVIDIA H100 NVL 94GB"
vram_gb: 94
region:
id: "REG0054"
name: "Virginia"
country: "United States"
requested_gpus: 16
terms:
deployment_type: "Virtual Machine"
contract_type: "On-Demand"
billing_model: "Hourly"
minimum_term_months: null
price:
original:
value: 6.98
currency: "USD"
unit: "instance-hour"
normalized_usd_per_gpu_hour: 6.98
availability:
status: "Unconfirmed"
recorded_gpu_count: null
evidence:
confidence_grade: "B"
last_checked: "2026-11-17T12:04:15.45+00:00"
Match 8
Microsoft Azure · H100 NVL 94GB
Virginia
Unconfirmed
$6.98 USD / GPU-hourTotal not estimated: add the GPU count and duration · Grade B
1 requirements need confirmation. Compute-only estimate; additional charges and final terms may apply.
Why this match · economics & evidence
GPU requirement matches the tracked accelerator.
Quantity compatibility cannot be confirmed from current capacity evidence.
Billed planning month is 730 hours unless the workload uses an explicit run duration. Utilization assumption is 70.0%. No egress volume was requested, so egress cost is excluded from the scenario. This scenario models compute commitments, requested egress, storage, setup, interconnect and entered overhead; taxes and unlisted charges require contract review. Minimum term is undisclosed; no complete contractual total is claimed. Billed run duration is not specified or derivable. Setup fee is undisclosed or invalid; it is not assumed to be zero. No applicable public comparable performance benchmark or explicit throughput assumption is available.
Opens the provider's own site. Reconfirm price and availability there before you commit.
Offering brief · take this to the provider
A sourced summary of this catalogue record: original price and unit, normalised price, confidence grade and check date. It is not a quote or a reservation.
# Kovara offering brief
# Generated 2026-11-17T14:33:26.998Z. Values are catalogue records, not a reservation or a quote.
# Reconfirm price and availability with the provider before you commit.
kovara:
offering_id: "OFF000146"
provider:
id: "PRV0001"
name: "Microsoft Azure"
url: "https://azure.microsoft.com/en-us/pricing/details/virtual-machines/linux/"
gpu:
id: "GPU0002"
name: "NVIDIA H100 NVL 94GB"
vram_gb: 94
region:
id: "REG0054"
name: "Virginia"
country: "United States"
requested_gpus: 16
terms:
deployment_type: "Virtual Machine"
contract_type: "On-Demand"
billing_model: "Hourly"
minimum_term_months: null
price:
original:
value: 6.98
currency: "USD"
unit: "instance-hour"
normalized_usd_per_gpu_hour: 6.98
availability:
status: "Unconfirmed"
recorded_gpu_count: null
evidence:
confidence_grade: "B"
last_checked: "2026-11-17T06:04:09.721+00:00"
Match 9
Microsoft Azure · H100 NVL 94GB
Washington State
Unconfirmed
$6.98 USD / GPU-hourTotal not estimated: add the GPU count and duration · Grade B
1 requirements need confirmation. Compute-only estimate; additional charges and final terms may apply.
Why this match · economics & evidence
GPU requirement matches the tracked accelerator.
Quantity compatibility cannot be confirmed from current capacity evidence.
Billed planning month is 730 hours unless the workload uses an explicit run duration. Utilization assumption is 70.0%. No egress volume was requested, so egress cost is excluded from the scenario. This scenario models compute commitments, requested egress, storage, setup, interconnect and entered overhead; taxes and unlisted charges require contract review. Minimum term is undisclosed; no complete contractual total is claimed. Billed run duration is not specified or derivable. Setup fee is undisclosed or invalid; it is not assumed to be zero. No applicable public comparable performance benchmark or explicit throughput assumption is available.
Opens the provider's own site. Reconfirm price and availability there before you commit.
Offering brief · take this to the provider
A sourced summary of this catalogue record: original price and unit, normalised price, confidence grade and check date. It is not a quote or a reservation.
# Kovara offering brief
# Generated 2026-11-17T14:33:26.998Z. Values are catalogue records, not a reservation or a quote.
# Reconfirm price and availability with the provider before you commit.
kovara:
offering_id: "OFF000165"
provider:
id: "PRV0001"
name: "Microsoft Azure"
url: "https://azure.microsoft.com/en-us/pricing/details/virtual-machines/linux/"
gpu:
id: "GPU0002"
name: "NVIDIA H100 NVL 94GB"
vram_gb: 94
region:
id: "REG0097"
name: "Washington State"
country: "United States"
requested_gpus: 16
terms:
deployment_type: "Virtual Machine"
contract_type: "On-Demand"
billing_model: "Hourly"
minimum_term_months: null
price:
original:
value: 6.98
currency: "USD"
unit: "instance-hour"
normalized_usd_per_gpu_hour: 6.98
availability:
status: "Unconfirmed"
recorded_gpu_count: null
evidence:
confidence_grade: "B"
last_checked: "2026-11-17T12:07:10.212+00:00"
Match 10
Microsoft Azure · H100 NVL 94GB
Phoenix
Unconfirmed
$6.98 USD / GPU-hourTotal not estimated: add the GPU count and duration · Grade B
1 requirements need confirmation. Compute-only estimate; additional charges and final terms may apply.
Why this match · economics & evidence
GPU requirement matches the tracked accelerator.
Quantity compatibility cannot be confirmed from current capacity evidence.
Billed planning month is 730 hours unless the workload uses an explicit run duration. Utilization assumption is 70.0%. No egress volume was requested, so egress cost is excluded from the scenario. This scenario models compute commitments, requested egress, storage, setup, interconnect and entered overhead; taxes and unlisted charges require contract review. Minimum term is undisclosed; no complete contractual total is claimed. Billed run duration is not specified or derivable. Setup fee is undisclosed or invalid; it is not assumed to be zero. No applicable public comparable performance benchmark or explicit throughput assumption is available.
Opens the provider's own site. Reconfirm price and availability there before you commit.
Offering brief · take this to the provider
A sourced summary of this catalogue record: original price and unit, normalised price, confidence grade and check date. It is not a quote or a reservation.
# Kovara offering brief
# Generated 2026-11-17T14:33:26.998Z. Values are catalogue records, not a reservation or a quote.
# Reconfirm price and availability with the provider before you commit.
kovara:
offering_id: "OFF000166"
provider:
id: "PRV0001"
name: "Microsoft Azure"
url: "https://azure.microsoft.com/en-us/pricing/details/virtual-machines/linux/"
gpu:
id: "GPU0002"
name: "NVIDIA H100 NVL 94GB"
vram_gb: 94
region:
id: "REG0010"
name: "Phoenix"
country: "United States"
requested_gpus: 16
terms:
deployment_type: "Virtual Machine"
contract_type: "On-Demand"
billing_model: "Hourly"
minimum_term_months: null
price:
original:
value: 6.98
currency: "USD"
unit: "instance-hour"
normalized_usd_per_gpu_hour: 6.98
availability:
status: "Unconfirmed"
recorded_gpu_count: null
evidence:
confidence_grade: "B"
last_checked: "2026-11-17T12:07:10.212+00:00"
Match 11
Microsoft Azure · H100 NVL 94GB
Virginia
Unconfirmed
$6.98 USD / GPU-hourTotal not estimated: add the GPU count and duration · Grade B
1 requirements need confirmation. Compute-only estimate; additional charges and final terms may apply.
Why this match · economics & evidence
GPU requirement matches the tracked accelerator.
Quantity compatibility cannot be confirmed from current capacity evidence.
Billed planning month is 730 hours unless the workload uses an explicit run duration. Utilization assumption is 70.0%. No egress volume was requested, so egress cost is excluded from the scenario. This scenario models compute commitments, requested egress, storage, setup, interconnect and entered overhead; taxes and unlisted charges require contract review. Minimum term is undisclosed; no complete contractual total is claimed. Billed run duration is not specified or derivable. Setup fee is undisclosed or invalid; it is not assumed to be zero. No applicable public comparable performance benchmark or explicit throughput assumption is available.
Opens the provider's own site. Reconfirm price and availability there before you commit.
Offering brief · take this to the provider
A sourced summary of this catalogue record: original price and unit, normalised price, confidence grade and check date. It is not a quote or a reservation.
# Kovara offering brief
# Generated 2026-11-17T14:33:26.998Z. Values are catalogue records, not a reservation or a quote.
# Reconfirm price and availability with the provider before you commit.
kovara:
offering_id: "OFF000254"
provider:
id: "PRV0001"
name: "Microsoft Azure"
url: "https://azure.microsoft.com/en-us/pricing/details/virtual-machines/linux/"
gpu:
id: "GPU0002"
name: "NVIDIA H100 NVL 94GB"
vram_gb: 94
region:
id: "REG0054"
name: "Virginia"
country: "United States"
requested_gpus: 16
terms:
deployment_type: "Virtual Machine"
contract_type: "On-Demand"
billing_model: "Hourly"
minimum_term_months: null
price:
original:
value: 13.96
currency: "USD"
unit: "instance-hour"
normalized_usd_per_gpu_hour: 6.98
availability:
status: "Unconfirmed"
recorded_gpu_count: null
evidence:
confidence_grade: "B"
last_checked: "2026-11-17T12:11:10.426+00:00"
Match 12
Microsoft Azure · H100 NVL 94GB
Virginia
Unconfirmed
$6.98 USD / GPU-hourTotal not estimated: add the GPU count and duration · Grade B
1 requirements need confirmation. Compute-only estimate; additional charges and final terms may apply.
Why this match · economics & evidence
GPU requirement matches the tracked accelerator.
Quantity compatibility cannot be confirmed from current capacity evidence.
Billed planning month is 730 hours unless the workload uses an explicit run duration. Utilization assumption is 70.0%. No egress volume was requested, so egress cost is excluded from the scenario. This scenario models compute commitments, requested egress, storage, setup, interconnect and entered overhead; taxes and unlisted charges require contract review. Minimum term is undisclosed; no complete contractual total is claimed. Billed run duration is not specified or derivable. Setup fee is undisclosed or invalid; it is not assumed to be zero. No applicable public comparable performance benchmark or explicit throughput assumption is available.
Opens the provider's own site. Reconfirm price and availability there before you commit.
Offering brief · take this to the provider
A sourced summary of this catalogue record: original price and unit, normalised price, confidence grade and check date. It is not a quote or a reservation.
# Kovara offering brief
# Generated 2026-11-17T14:33:26.998Z. Values are catalogue records, not a reservation or a quote.
# Reconfirm price and availability with the provider before you commit.
kovara:
offering_id: "OFF000255"
provider:
id: "PRV0001"
name: "Microsoft Azure"
url: "https://azure.microsoft.com/en-us/pricing/details/virtual-machines/linux/"
gpu:
id: "GPU0002"
name: "NVIDIA H100 NVL 94GB"
vram_gb: 94
region:
id: "REG0054"
name: "Virginia"
country: "United States"
requested_gpus: 16
terms:
deployment_type: "Virtual Machine"
contract_type: "On-Demand"
billing_model: "Hourly"
minimum_term_months: null
price:
original:
value: 13.96
currency: "USD"
unit: "instance-hour"
normalized_usd_per_gpu_hour: 6.98
availability:
status: "Unconfirmed"
recorded_gpu_count: null
evidence:
confidence_grade: "B"
last_checked: "2026-11-17T12:11:10.426+00:00"
Match 13
Microsoft Azure · H100 NVL 94GB
Washington State
Unconfirmed
$6.98 USD / GPU-hourTotal not estimated: add the GPU count and duration · Grade B
1 requirements need confirmation. Compute-only estimate; additional charges and final terms may apply.
Why this match · economics & evidence
GPU requirement matches the tracked accelerator.
Quantity compatibility cannot be confirmed from current capacity evidence.
Billed planning month is 730 hours unless the workload uses an explicit run duration. Utilization assumption is 70.0%. No egress volume was requested, so egress cost is excluded from the scenario. This scenario models compute commitments, requested egress, storage, setup, interconnect and entered overhead; taxes and unlisted charges require contract review. Minimum term is undisclosed; no complete contractual total is claimed. Billed run duration is not specified or derivable. Setup fee is undisclosed or invalid; it is not assumed to be zero. No applicable public comparable performance benchmark or explicit throughput assumption is available.
Opens the provider's own site. Reconfirm price and availability there before you commit.
Offering brief · take this to the provider
A sourced summary of this catalogue record: original price and unit, normalised price, confidence grade and check date. It is not a quote or a reservation.
# Kovara offering brief
# Generated 2026-11-17T14:33:26.998Z. Values are catalogue records, not a reservation or a quote.
# Reconfirm price and availability with the provider before you commit.
kovara:
offering_id: "OFF000274"
provider:
id: "PRV0001"
name: "Microsoft Azure"
url: "https://azure.microsoft.com/en-us/pricing/details/virtual-machines/linux/"
gpu:
id: "GPU0002"
name: "NVIDIA H100 NVL 94GB"
vram_gb: 94
region:
id: "REG0097"
name: "Washington State"
country: "United States"
requested_gpus: 16
terms:
deployment_type: "Virtual Machine"
contract_type: "On-Demand"
billing_model: "Hourly"
minimum_term_months: null
price:
original:
value: 13.96
currency: "USD"
unit: "instance-hour"
normalized_usd_per_gpu_hour: 6.98
availability:
status: "Unconfirmed"
recorded_gpu_count: null
evidence:
confidence_grade: "B"
last_checked: "2026-11-17T12:14:10.068+00:00"
Match 14
Microsoft Azure · H100 NVL 94GB
Phoenix
Unconfirmed
$6.98 USD / GPU-hourTotal not estimated: add the GPU count and duration · Grade B
1 requirements need confirmation. Compute-only estimate; additional charges and final terms may apply.
Why this match · economics & evidence
GPU requirement matches the tracked accelerator.
Quantity compatibility cannot be confirmed from current capacity evidence.
Billed planning month is 730 hours unless the workload uses an explicit run duration. Utilization assumption is 70.0%. No egress volume was requested, so egress cost is excluded from the scenario. This scenario models compute commitments, requested egress, storage, setup, interconnect and entered overhead; taxes and unlisted charges require contract review. Minimum term is undisclosed; no complete contractual total is claimed. Billed run duration is not specified or derivable. Setup fee is undisclosed or invalid; it is not assumed to be zero. No applicable public comparable performance benchmark or explicit throughput assumption is available.
Opens the provider's own site. Reconfirm price and availability there before you commit.
Offering brief · take this to the provider
A sourced summary of this catalogue record: original price and unit, normalised price, confidence grade and check date. It is not a quote or a reservation.
# Kovara offering brief
# Generated 2026-11-17T14:33:26.998Z. Values are catalogue records, not a reservation or a quote.
# Reconfirm price and availability with the provider before you commit.
kovara:
offering_id: "OFF000275"
provider:
id: "PRV0001"
name: "Microsoft Azure"
url: "https://azure.microsoft.com/en-us/pricing/details/virtual-machines/linux/"
gpu:
id: "GPU0002"
name: "NVIDIA H100 NVL 94GB"
vram_gb: 94
region:
id: "REG0010"
name: "Phoenix"
country: "United States"
requested_gpus: 16
terms:
deployment_type: "Virtual Machine"
contract_type: "On-Demand"
billing_model: "Hourly"
minimum_term_months: null
price:
original:
value: 13.96
currency: "USD"
unit: "instance-hour"
normalized_usd_per_gpu_hour: 6.98
availability:
status: "Unconfirmed"
recorded_gpu_count: null
evidence:
confidence_grade: "B"
last_checked: "2026-11-17T12:14:10.068+00:00"
Match 15
Microsoft Azure · H100 NVL 94GB
Illinois
Unconfirmed
$8.38 USD / GPU-hourTotal not estimated: add the GPU count and duration · Grade B
1 requirements need confirmation. Compute-only estimate; additional charges and final terms may apply.
Why this match · economics & evidence
GPU requirement matches the tracked accelerator.
Quantity compatibility cannot be confirmed from current capacity evidence.
Billed planning month is 730 hours unless the workload uses an explicit run duration. Utilization assumption is 70.0%. No egress volume was requested, so egress cost is excluded from the scenario. This scenario models compute commitments, requested egress, storage, setup, interconnect and entered overhead; taxes and unlisted charges require contract review. Minimum term is undisclosed; no complete contractual total is claimed. Billed run duration is not specified or derivable. Setup fee is undisclosed or invalid; it is not assumed to be zero. No applicable public comparable performance benchmark or explicit throughput assumption is available.
Opens the provider's own site. Reconfirm price and availability there before you commit.
Offering brief · take this to the provider
A sourced summary of this catalogue record: original price and unit, normalised price, confidence grade and check date. It is not a quote or a reservation.
# Kovara offering brief
# Generated 2026-11-17T14:33:26.998Z. Values are catalogue records, not a reservation or a quote.
# Reconfirm price and availability with the provider before you commit.
kovara:
offering_id: "OFF000153"
provider:
id: "PRV0001"
name: "Microsoft Azure"
url: "https://azure.microsoft.com/en-us/pricing/details/virtual-machines/linux/"
gpu:
id: "GPU0002"
name: "NVIDIA H100 NVL 94GB"
vram_gb: 94
region:
id: "REG0090"
name: "Illinois"
country: "United States"
requested_gpus: 16
terms:
deployment_type: "Virtual Machine"
contract_type: "On-Demand"
billing_model: "Hourly"
minimum_term_months: null
price:
original:
value: 8.376
currency: "USD"
unit: "instance-hour"
normalized_usd_per_gpu_hour: 8.376
availability:
status: "Unconfirmed"
recorded_gpu_count: null
evidence:
confidence_grade: "B"
last_checked: "2026-11-17T12:05:13.08+00:00"
Match 16
Microsoft Azure · H100 NVL 94GB
Illinois
Unconfirmed
$8.38 USD / GPU-hourTotal not estimated: add the GPU count and duration · Grade B
1 requirements need confirmation. Compute-only estimate; additional charges and final terms may apply.
Why this match · economics & evidence
GPU requirement matches the tracked accelerator.
Quantity compatibility cannot be confirmed from current capacity evidence.
Billed planning month is 730 hours unless the workload uses an explicit run duration. Utilization assumption is 70.0%. No egress volume was requested, so egress cost is excluded from the scenario. This scenario models compute commitments, requested egress, storage, setup, interconnect and entered overhead; taxes and unlisted charges require contract review. Minimum term is undisclosed; no complete contractual total is claimed. Billed run duration is not specified or derivable. Setup fee is undisclosed or invalid; it is not assumed to be zero. No applicable public comparable performance benchmark or explicit throughput assumption is available.
Opens the provider's own site. Reconfirm price and availability there before you commit.
Offering brief · take this to the provider
A sourced summary of this catalogue record: original price and unit, normalised price, confidence grade and check date. It is not a quote or a reservation.
# Kovara offering brief
# Generated 2026-11-17T14:33:26.998Z. Values are catalogue records, not a reservation or a quote.
# Reconfirm price and availability with the provider before you commit.
kovara:
offering_id: "OFF000262"
provider:
id: "PRV0001"
name: "Microsoft Azure"
url: "https://azure.microsoft.com/en-us/pricing/details/virtual-machines/linux/"
gpu:
id: "GPU0002"
name: "NVIDIA H100 NVL 94GB"
vram_gb: 94
region:
id: "REG0090"
name: "Illinois"
country: "United States"
requested_gpus: 16
terms:
deployment_type: "Virtual Machine"
contract_type: "On-Demand"
billing_model: "Hourly"
minimum_term_months: null
price:
original:
value: 16.752
currency: "USD"
unit: "instance-hour"
normalized_usd_per_gpu_hour: 8.376
availability:
status: "Unconfirmed"
recorded_gpu_count: null
evidence:
confidence_grade: "B"
last_checked: "2026-11-17T12:12:25.527+00:00"
Match 17
Microsoft Azure · H100 NVL 94GB
Texas
Unconfirmed
$8.38 USD / GPU-hourTotal not estimated: add the GPU count and duration · Grade B
1 requirements need confirmation. Compute-only estimate; additional charges and final terms may apply.
Why this match · economics & evidence
GPU requirement matches the tracked accelerator.
Quantity compatibility cannot be confirmed from current capacity evidence.
Billed planning month is 730 hours unless the workload uses an explicit run duration. Utilization assumption is 70.0%. No egress volume was requested, so egress cost is excluded from the scenario. This scenario models compute commitments, requested egress, storage, setup, interconnect and entered overhead; taxes and unlisted charges require contract review. Minimum term is undisclosed; no complete contractual total is claimed. Billed run duration is not specified or derivable. Setup fee is undisclosed or invalid; it is not assumed to be zero. No applicable public comparable performance benchmark or explicit throughput assumption is available.
Opens the provider's own site. Reconfirm price and availability there before you commit.
Offering brief · take this to the provider
A sourced summary of this catalogue record: original price and unit, normalised price, confidence grade and check date. It is not a quote or a reservation.
# Kovara offering brief
# Generated 2026-11-17T14:33:26.998Z. Values are catalogue records, not a reservation or a quote.
# Reconfirm price and availability with the provider before you commit.
kovara:
offering_id: "OFF000155"
provider:
id: "PRV0001"
name: "Microsoft Azure"
url: "https://azure.microsoft.com/en-us/pricing/details/virtual-machines/linux/"
gpu:
id: "GPU0002"
name: "NVIDIA H100 NVL 94GB"
vram_gb: 94
region:
id: "REG0004"
name: "Texas"
country: "United States"
requested_gpus: 16
terms:
deployment_type: "Virtual Machine"
contract_type: "On-Demand"
billing_model: "Hourly"
minimum_term_months: null
price:
original:
value: 8.38
currency: "USD"
unit: "instance-hour"
normalized_usd_per_gpu_hour: 8.38
availability:
status: "Unconfirmed"
recorded_gpu_count: null
evidence:
confidence_grade: "B"
last_checked: "2026-11-17T12:06:09.847+00:00"
Match 18
Microsoft Azure · H100 NVL 94GB
Texas
Unconfirmed
$8.38 USD / GPU-hourTotal not estimated: add the GPU count and duration · Grade B
1 requirements need confirmation. Compute-only estimate; additional charges and final terms may apply.
Why this match · economics & evidence
GPU requirement matches the tracked accelerator.
Quantity compatibility cannot be confirmed from current capacity evidence.
Billed planning month is 730 hours unless the workload uses an explicit run duration. Utilization assumption is 70.0%. No egress volume was requested, so egress cost is excluded from the scenario. This scenario models compute commitments, requested egress, storage, setup, interconnect and entered overhead; taxes and unlisted charges require contract review. Minimum term is undisclosed; no complete contractual total is claimed. Billed run duration is not specified or derivable. Setup fee is undisclosed or invalid; it is not assumed to be zero. No applicable public comparable performance benchmark or explicit throughput assumption is available.
Opens the provider's own site. Reconfirm price and availability there before you commit.
Offering brief · take this to the provider
A sourced summary of this catalogue record: original price and unit, normalised price, confidence grade and check date. It is not a quote or a reservation.
# Kovara offering brief
# Generated 2026-11-17T14:33:26.998Z. Values are catalogue records, not a reservation or a quote.
# Reconfirm price and availability with the provider before you commit.
kovara:
offering_id: "OFF000264"
provider:
id: "PRV0001"
name: "Microsoft Azure"
url: "https://azure.microsoft.com/en-us/pricing/details/virtual-machines/linux/"
gpu:
id: "GPU0002"
name: "NVIDIA H100 NVL 94GB"
vram_gb: 94
region:
id: "REG0004"
name: "Texas"
country: "United States"
requested_gpus: 16
terms:
deployment_type: "Virtual Machine"
contract_type: "On-Demand"
billing_model: "Hourly"
minimum_term_months: null
price:
original:
value: 16.76
currency: "USD"
unit: "instance-hour"
normalized_usd_per_gpu_hour: 8.38
availability:
status: "Unconfirmed"
recorded_gpu_count: null
evidence:
confidence_grade: "B"
last_checked: "2026-11-17T12:13:09.918+00:00"
Match 19
Microsoft Azure · H100 NVL 94GB
Iowa
Unconfirmed
$8.59 USD / GPU-hourTotal not estimated: add the GPU count and duration · Grade B
1 requirements need confirmation. Compute-only estimate; additional charges and final terms may apply.
Why this match · economics & evidence
GPU requirement matches the tracked accelerator.
Quantity compatibility cannot be confirmed from current capacity evidence.
Billed planning month is 730 hours unless the workload uses an explicit run duration. Utilization assumption is 70.0%. No egress volume was requested, so egress cost is excluded from the scenario. This scenario models compute commitments, requested egress, storage, setup, interconnect and entered overhead; taxes and unlisted charges require contract review. Minimum term is undisclosed; no complete contractual total is claimed. Billed run duration is not specified or derivable. Setup fee is undisclosed or invalid; it is not assumed to be zero. No applicable public comparable performance benchmark or explicit throughput assumption is available.
Opens the provider's own site. Reconfirm price and availability there before you commit.
Offering brief · take this to the provider
A sourced summary of this catalogue record: original price and unit, normalised price, confidence grade and check date. It is not a quote or a reservation.
# Kovara offering brief
# Generated 2026-11-17T14:33:26.998Z. Values are catalogue records, not a reservation or a quote.
# Reconfirm price and availability with the provider before you commit.
kovara:
offering_id: "OFF000144"
provider:
id: "PRV0001"
name: "Microsoft Azure"
url: "https://azure.microsoft.com/en-us/pricing/details/virtual-machines/linux/"
gpu:
id: "GPU0002"
name: "NVIDIA H100 NVL 94GB"
vram_gb: 94
region:
id: "REG0055"
name: "Iowa"
country: "United States"
requested_gpus: 16
terms:
deployment_type: "Virtual Machine"
contract_type: "On-Demand"
billing_model: "Hourly"
minimum_term_months: null
price:
original:
value: 8.585
currency: "USD"
unit: "instance-hour"
normalized_usd_per_gpu_hour: 8.585
availability:
status: "Unconfirmed"
recorded_gpu_count: null
evidence:
confidence_grade: "B"
last_checked: "2026-11-17T12:04:15.45+00:00"
Match 20
Microsoft Azure · H100 NVL 94GB
Iowa
Unconfirmed
$8.59 USD / GPU-hourTotal not estimated: add the GPU count and duration · Grade B
1 requirements need confirmation. Compute-only estimate; additional charges and final terms may apply.
Why this match · economics & evidence
GPU requirement matches the tracked accelerator.
Quantity compatibility cannot be confirmed from current capacity evidence.
Billed planning month is 730 hours unless the workload uses an explicit run duration. Utilization assumption is 70.0%. No egress volume was requested, so egress cost is excluded from the scenario. This scenario models compute commitments, requested egress, storage, setup, interconnect and entered overhead; taxes and unlisted charges require contract review. Minimum term is undisclosed; no complete contractual total is claimed. Billed run duration is not specified or derivable. Setup fee is undisclosed or invalid; it is not assumed to be zero. No applicable public comparable performance benchmark or explicit throughput assumption is available.
Opens the provider's own site. Reconfirm price and availability there before you commit.
Offering brief · take this to the provider
A sourced summary of this catalogue record: original price and unit, normalised price, confidence grade and check date. It is not a quote or a reservation.
# Kovara offering brief
# Generated 2026-11-17T14:33:26.998Z. Values are catalogue records, not a reservation or a quote.
# Reconfirm price and availability with the provider before you commit.
kovara:
offering_id: "OFF000253"
provider:
id: "PRV0001"
name: "Microsoft Azure"
url: "https://azure.microsoft.com/en-us/pricing/details/virtual-machines/linux/"
gpu:
id: "GPU0002"
name: "NVIDIA H100 NVL 94GB"
vram_gb: 94
region:
id: "REG0055"
name: "Iowa"
country: "United States"
requested_gpus: 16
terms:
deployment_type: "Virtual Machine"
contract_type: "On-Demand"
billing_model: "Hourly"
minimum_term_months: null
price:
original:
value: 17.17
currency: "USD"
unit: "instance-hour"
normalized_usd_per_gpu_hour: 8.585
availability:
status: "Unconfirmed"
recorded_gpu_count: null
evidence:
confidence_grade: "B"
last_checked: "2026-11-17T12:10:41.164+00:00"
Match 21
Microsoft Azure · H100 NVL 94GB
Toronto
Unconfirmed
$8.59 USD / GPU-hourTotal not estimated: add the GPU count and duration · Grade B
1 requirements need confirmation. Compute-only estimate; additional charges and final terms may apply.
Why this match · economics & evidence
GPU requirement matches the tracked accelerator.
Quantity compatibility cannot be confirmed from current capacity evidence.
Billed planning month is 730 hours unless the workload uses an explicit run duration. Utilization assumption is 70.0%. No egress volume was requested, so egress cost is excluded from the scenario. This scenario models compute commitments, requested egress, storage, setup, interconnect and entered overhead; taxes and unlisted charges require contract review. Minimum term is undisclosed; no complete contractual total is claimed. Billed run duration is not specified or derivable. Setup fee is undisclosed or invalid; it is not assumed to be zero. No applicable public comparable performance benchmark or explicit throughput assumption is available.
Opens the provider's own site. Reconfirm price and availability there before you commit.
Offering brief · take this to the provider
A sourced summary of this catalogue record: original price and unit, normalised price, confidence grade and check date. It is not a quote or a reservation.
# Kovara offering brief
# Generated 2026-11-17T14:33:26.998Z. Values are catalogue records, not a reservation or a quote.
# Reconfirm price and availability with the provider before you commit.
kovara:
offering_id: "OFF000142"
provider:
id: "PRV0001"
name: "Microsoft Azure"
url: "https://azure.microsoft.com/en-us/pricing/details/virtual-machines/linux/"
gpu:
id: "GPU0002"
name: "NVIDIA H100 NVL 94GB"
vram_gb: 94
region:
id: "REG0014"
name: "Toronto"
country: "Canada"
requested_gpus: 16
terms:
deployment_type: "Virtual Machine"
contract_type: "On-Demand"
billing_model: "Hourly"
minimum_term_months: null
price:
original:
value: 8.59
currency: "USD"
unit: "instance-hour"
normalized_usd_per_gpu_hour: 8.59
availability:
status: "Unconfirmed"
recorded_gpu_count: null
evidence:
confidence_grade: "B"
last_checked: "2026-11-17T12:04:15.45+00:00"
Match 22
Microsoft Azure · H100 NVL 94GB
Toronto
Unconfirmed
$8.59 USD / GPU-hourTotal not estimated: add the GPU count and duration · Grade B
1 requirements need confirmation. Compute-only estimate; additional charges and final terms may apply.
Why this match · economics & evidence
GPU requirement matches the tracked accelerator.
Quantity compatibility cannot be confirmed from current capacity evidence.
Billed planning month is 730 hours unless the workload uses an explicit run duration. Utilization assumption is 70.0%. No egress volume was requested, so egress cost is excluded from the scenario. This scenario models compute commitments, requested egress, storage, setup, interconnect and entered overhead; taxes and unlisted charges require contract review. Minimum term is undisclosed; no complete contractual total is claimed. Billed run duration is not specified or derivable. Setup fee is undisclosed or invalid; it is not assumed to be zero. No applicable public comparable performance benchmark or explicit throughput assumption is available.
Opens the provider's own site. Reconfirm price and availability there before you commit.
Offering brief · take this to the provider
A sourced summary of this catalogue record: original price and unit, normalised price, confidence grade and check date. It is not a quote or a reservation.
# Kovara offering brief
# Generated 2026-11-17T14:33:26.998Z. Values are catalogue records, not a reservation or a quote.
# Reconfirm price and availability with the provider before you commit.
kovara:
offering_id: "OFF000251"
provider:
id: "PRV0001"
name: "Microsoft Azure"
url: "https://azure.microsoft.com/en-us/pricing/details/virtual-machines/linux/"
gpu:
id: "GPU0002"
name: "NVIDIA H100 NVL 94GB"
vram_gb: 94
region:
id: "REG0014"
name: "Toronto"
country: "Canada"
requested_gpus: 16
terms:
deployment_type: "Virtual Machine"
contract_type: "On-Demand"
billing_model: "Hourly"
minimum_term_months: null
price:
original:
value: 17.18
currency: "USD"
unit: "instance-hour"
normalized_usd_per_gpu_hour: 8.59
availability:
status: "Unconfirmed"
recorded_gpu_count: null
evidence:
confidence_grade: "B"
last_checked: "2026-11-17T06:09:25.819+00:00"
Match 23
Microsoft Azure · H100 NVL 94GB
California
Unconfirmed
$9.07 USD / GPU-hourTotal not estimated: add the GPU count and duration · Grade B
1 requirements need confirmation. Compute-only estimate; additional charges and final terms may apply.
Why this match · economics & evidence
GPU requirement matches the tracked accelerator.
Quantity compatibility cannot be confirmed from current capacity evidence.
Billed planning month is 730 hours unless the workload uses an explicit run duration. Utilization assumption is 70.0%. No egress volume was requested, so egress cost is excluded from the scenario. This scenario models compute commitments, requested egress, storage, setup, interconnect and entered overhead; taxes and unlisted charges require contract review. Minimum term is undisclosed; no complete contractual total is claimed. Billed run duration is not specified or derivable. Setup fee is undisclosed or invalid; it is not assumed to be zero. No applicable public comparable performance benchmark or explicit throughput assumption is available.
Opens the provider's own site. Reconfirm price and availability there before you commit.
Offering brief · take this to the provider
A sourced summary of this catalogue record: original price and unit, normalised price, confidence grade and check date. It is not a quote or a reservation.
# Kovara offering brief
# Generated 2026-11-17T14:33:26.998Z. Values are catalogue records, not a reservation or a quote.
# Reconfirm price and availability with the provider before you commit.
kovara:
offering_id: "OFF000164"
provider:
id: "PRV0001"
name: "Microsoft Azure"
url: "https://azure.microsoft.com/en-us/pricing/details/virtual-machines/linux/"
gpu:
id: "GPU0002"
name: "NVIDIA H100 NVL 94GB"
vram_gb: 94
region:
id: "REG0096"
name: "California"
country: "United States"
requested_gpus: 16
terms:
deployment_type: "Virtual Machine"
contract_type: "On-Demand"
billing_model: "Hourly"
minimum_term_months: null
price:
original:
value: 9.074
currency: "USD"
unit: "instance-hour"
normalized_usd_per_gpu_hour: 9.074
availability:
status: "Unconfirmed"
recorded_gpu_count: null
evidence:
confidence_grade: "B"
last_checked: "2026-11-17T12:07:10.212+00:00"
Match 24
Microsoft Azure · H100 NVL 94GB
California
Unconfirmed
$9.07 USD / GPU-hourTotal not estimated: add the GPU count and duration · Grade B
1 requirements need confirmation. Compute-only estimate; additional charges and final terms may apply.
Why this match · economics & evidence
GPU requirement matches the tracked accelerator.
Quantity compatibility cannot be confirmed from current capacity evidence.
Billed planning month is 730 hours unless the workload uses an explicit run duration. Utilization assumption is 70.0%. No egress volume was requested, so egress cost is excluded from the scenario. This scenario models compute commitments, requested egress, storage, setup, interconnect and entered overhead; taxes and unlisted charges require contract review. Minimum term is undisclosed; no complete contractual total is claimed. Billed run duration is not specified or derivable. Setup fee is undisclosed or invalid; it is not assumed to be zero. No applicable public comparable performance benchmark or explicit throughput assumption is available.
Opens the provider's own site. Reconfirm price and availability there before you commit.
Offering brief · take this to the provider
A sourced summary of this catalogue record: original price and unit, normalised price, confidence grade and check date. It is not a quote or a reservation.
# Kovara offering brief
# Generated 2026-11-17T14:33:26.998Z. Values are catalogue records, not a reservation or a quote.
# Reconfirm price and availability with the provider before you commit.
kovara:
offering_id: "OFF000273"
provider:
id: "PRV0001"
name: "Microsoft Azure"
url: "https://azure.microsoft.com/en-us/pricing/details/virtual-machines/linux/"
gpu:
id: "GPU0002"
name: "NVIDIA H100 NVL 94GB"
vram_gb: 94
region:
id: "REG0096"
name: "California"
country: "United States"
requested_gpus: 16
terms:
deployment_type: "Virtual Machine"
contract_type: "On-Demand"
billing_model: "Hourly"
minimum_term_months: null
price:
original:
value: 18.148
currency: "USD"
unit: "instance-hour"
normalized_usd_per_gpu_hour: 9.074
availability:
status: "Unconfirmed"
recorded_gpu_count: null
evidence:
confidence_grade: "B"
last_checked: "2026-11-17T12:14:10.068+00:00"
Showing the 24 highest-ranked configurations. Refine your requirement to narrow the results.