Data interface · Agent and API manual

Cloud GPU Pricing Catalog

Filter accepted current offerings without treating incomplete component prices as totals. Every result retains its source, verification time, geography state, and comparison eligibility.

Reset

Hourly price sorting intentionally requires “complete comparable totals” and exactly one currency. Repeat parameters such as provider are supported by the API and URL contract.

Results

136 matching offerings · 212 commercial variants

Provider Offering Commercial variant Price EU/EEA Evidence
Baseten.co
Dedicated Deployments
A10G
A10G
fixed · GPU count unknown
Standard
scale to zero · standard · dedicated
inclusive total · not a complete comparable price
unknown gpu count
  • USD 0.02012/minute · compute time
Compute available
unknown
Constrain compute
unknown
EU-only service
unknown
current
Verified 13 Jul 2026
Official price source
Baseten.co
Dedicated Deployments
T4
T4
fixed · GPU count unknown
Standard
scale to zero · standard · dedicated
inclusive total · not a complete comparable price
unknown gpu count
  • USD 0.01052/minute · compute time
Compute available
unknown
Constrain compute
unknown
EU-only service
unknown
current
Verified 13 Jul 2026
Official price source
Baseten.co
Dedicated Deployments
L4
L4
fixed · GPU count unknown
Standard
scale to zero · standard · dedicated
inclusive total · not a complete comparable price
unknown gpu count
  • USD 0.01414/minute · compute time
Compute available
unknown
Constrain compute
unknown
EU-only service
unknown
current
Verified 13 Jul 2026
Official price source
Baseten.co
Dedicated Deployments
B200
NVIDIA B200
fixed · GPU count unknown
Standard
scale to zero · standard · dedicated
inclusive total · not a complete comparable price
unknown gpu count
  • USD 0.16633/minute · compute time
Compute available
unknown
Constrain compute
unknown
EU-only service
unknown
current
Verified 13 Jul 2026
Official price source
Baseten.co
Dedicated Deployments
H100 MIG
H100 MIG
opaque provider tier · GPU count unknown
Standard
scale to zero · standard · dedicated
inclusive total · not a complete comparable price
opaque hardware
  • USD 0.0625/minute · compute time
Compute available
unknown
Constrain compute
unknown
EU-only service
unknown
current
Verified 13 Jul 2026
Official price source
Baseten.co
Dedicated Deployments
H100
H100
fixed · GPU count unknown
Standard
scale to zero · standard · dedicated
inclusive total · not a complete comparable price
unknown gpu count
  • USD 0.10833/minute · compute time
Compute available
unknown
Constrain compute
unknown
EU-only service
unknown
current
Verified 13 Jul 2026
Official price source
Baseten.co
Dedicated Deployments
A100
A100
fixed · GPU count unknown
Standard
scale to zero · standard · dedicated
inclusive total · not a complete comparable price
unknown gpu count
  • USD 0.06667/minute · compute time
Compute available
unknown
Constrain compute
unknown
EU-only service
unknown
current
Verified 13 Jul 2026
Official price source
Beam.cloud
Serverless
B200
NVIDIA B200
fixed · 1 GPU
Standard
scale to zero · standard · unspecified
additive components · not a complete comparable price
additive price not total
  • USD 0.045/core hour · cpu core time
  • USD 5.62/hour · gpu time
  • USD 0.0162/gb hour · memory time
Compute available
unknown
Constrain compute
unknown
EU-only service
unknown
current
Verified 13 Jul 2026
Official price source
unpriced custom plan: The Serverless section shows a 'With committed spend' / 'Book a call' option without a published amount; it was omitted.
Beam.cloud
Serverless
RTX 5090
NVIDIA GeForce RTX 5090
fixed · 1 GPU
Standard
scale to zero · standard · unspecified
additive components · not a complete comparable price
additive price not total
  • USD 0.045/core hour · cpu core time
  • USD 1.09/hour · gpu time
  • USD 0.0162/gb hour · memory time
Compute available
unknown
Constrain compute
unknown
EU-only service
unknown
current
Verified 13 Jul 2026
Official price source
unpriced custom plan: The Serverless section shows a 'With committed spend' / 'Book a call' option without a published amount; it was omitted.
Beam.cloud
Serverless
L40S
L40S
fixed · 1 GPU
Standard
scale to zero · standard · unspecified
additive components · not a complete comparable price
additive price not total
  • USD 0.045/core hour · cpu core time
  • USD 1.75/hour · gpu time
  • USD 0.0162/gb hour · memory time
Compute available
unknown
Constrain compute
unknown
EU-only service
unknown
current
Verified 13 Jul 2026
Official price source
unpriced custom plan: The Serverless section shows a 'With committed spend' / 'Book a call' option without a published amount; it was omitted.
Beam.cloud
Serverless
A100 80GB
A100
fixed · 1 GPU
Standard
scale to zero · standard · unspecified
additive components · not a complete comparable price
additive price not total
  • USD 0.045/core hour · cpu core time
  • USD 2.25/hour · gpu time
  • USD 0.0162/gb hour · memory time
Compute available
unknown
Constrain compute
unknown
EU-only service
unknown
current
Verified 13 Jul 2026
Official price source
unpriced custom plan: The Serverless section shows a 'With committed spend' / 'Book a call' option without a published amount; it was omitted.
Beam.cloud
Serverless
H100
H100
fixed · 1 GPU
Standard
scale to zero · standard · unspecified
additive components · not a complete comparable price
additive price not total
  • USD 0.045/core hour · cpu core time
  • USD 3.55/hour · gpu time
  • USD 0.0162/gb hour · memory time
Compute available
unknown
Constrain compute
unknown
EU-only service
unknown
current
Verified 13 Jul 2026
Official price source
unpriced custom plan: The Serverless section shows a 'With committed spend' / 'Book a call' option without a published amount; it was omitted.
Beam.cloud
Serverless
RTX PRO 6000
NVIDIA RTX PRO 6000 Blackwell
fixed · 1 GPU
Standard
scale to zero · standard · unspecified
additive components · not a complete comparable price
additive price not total
  • USD 0.045/core hour · cpu core time
  • USD 2.73/hour · gpu time
  • USD 0.0162/gb hour · memory time
Compute available
unknown
Constrain compute
unknown
EU-only service
unknown
current
Verified 13 Jul 2026
Official price source
unpriced custom plan: The Serverless section shows a 'With committed spend' / 'Book a call' option without a published amount; it was omitted.
Beam.cloud
Serverless
RTX 4090
RTX 4090
fixed · 1 GPU
Standard
scale to zero · standard · unspecified
additive components · not a complete comparable price
additive price not total
  • USD 0.045/core hour · cpu core time
  • USD 0.69/hour · gpu time
  • USD 0.0162/gb hour · memory time
Compute available
unknown
Constrain compute
unknown
EU-only service
unknown
current
Verified 13 Jul 2026
Official price source
unpriced custom plan: The Serverless section shows a 'With committed spend' / 'Book a call' option without a published amount; it was omitted.
Beam.cloud
Serverless
H200
NVIDIA H200
fixed · 1 GPU
Standard
scale to zero · standard · unspecified
additive components · not a complete comparable price
additive price not total
  • USD 0.045/core hour · cpu core time
  • USD 4.09/hour · gpu time
  • USD 0.0162/gb hour · memory time
Compute available
unknown
Constrain compute
unknown
EU-only service
unknown
current
Verified 13 Jul 2026
Official price source
unpriced custom plan: The Serverless section shows a 'With committed spend' / 'Book a call' option without a published amount; it was omitted.
Beam.cloud
Serverless
A6000
A6000
fixed · 1 GPU
Standard
scale to zero · standard · unspecified
additive components · not a complete comparable price
additive price not total
  • USD 0.045/core hour · cpu core time
  • USD 0.82/hour · gpu time
  • USD 0.0162/gb hour · memory time
Compute available
unknown
Constrain compute
unknown
EU-only service
unknown
current
Verified 13 Jul 2026
Official price source
unpriced custom plan: The Serverless section shows a 'With committed spend' / 'Book a call' option without a published amount; it was omitted.
Cerebrium
Serverless
L4
L4
fixed · 1 GPU
Standard
scale to zero · standard · unspecified
additive components · not a complete comparable price
additive price not total
  • USD 0.000007/core second · cpu core time
  • USD 0.000222/second · gpu time
  • USD 0.000002/gb second · memory time
Compute available
unknown
Constrain compute
unknown
EU-only service
unknown
current
Verified 13 Jul 2026
Official price source
Cerebrium
Serverless
RTX PRO 6000
NVIDIA RTX PRO 6000 Blackwell
fixed · 1 GPU
Standard
scale to zero · standard · unspecified
additive components · not a complete comparable price
additive price not total
  • USD 0.000007/core second · cpu core time
  • USD 0.000694/second · gpu time
  • USD 0.000002/gb second · memory time
Compute available
unknown
Constrain compute
unknown
EU-only service
unknown
current
Verified 13 Jul 2026
Official price source
Cerebrium
Serverless
B200
NVIDIA B200
fixed · 1 GPU
Standard
scale to zero · standard · unspecified
additive components · not a complete comparable price
additive price not total
  • USD 0.000007/core second · cpu core time
  • USD 0.00167/second · gpu time
  • USD 0.000002/gb second · memory time
Compute available
unknown
Constrain compute
unknown
EU-only service
unknown
current
Verified 13 Jul 2026
Official price source
Cerebrium
Serverless
H100
H100
fixed · 1 GPU
Standard
scale to zero · standard · unspecified
additive components · not a complete comparable price
additive price not total
  • USD 0.000007/core second · cpu core time
  • USD 0.000944/second · gpu time
  • USD 0.000002/gb second · memory time
Compute available
unknown
Constrain compute
unknown
EU-only service
unknown
current
Verified 13 Jul 2026
Official price source
Cerebrium
Serverless
H200
NVIDIA H200
fixed · 1 GPU
Standard
scale to zero · standard · unspecified
additive components · not a complete comparable price
additive price not total
  • USD 0.000007/core second · cpu core time
  • USD 0.001166/second · gpu time
  • USD 0.000002/gb second · memory time
Compute available
unknown
Constrain compute
unknown
EU-only service
unknown
current
Verified 13 Jul 2026
Official price source
Cerebrium
Serverless
A100 (40GB)
A100
fixed · 1 GPU
Standard
scale to zero · standard · unspecified
additive components · not a complete comparable price
additive price not total
  • USD 0.000007/core second · cpu core time
  • USD 0.000555/second · gpu time
  • USD 0.000002/gb second · memory time
Compute available
unknown
Constrain compute
unknown
EU-only service
unknown
current
Verified 13 Jul 2026
Official price source
Cerebrium
Serverless
A10
A10
fixed · 1 GPU
Standard
scale to zero · standard · unspecified
additive components · not a complete comparable price
additive price not total
  • USD 0.000007/core second · cpu core time
  • USD 0.000306/second · gpu time
  • USD 0.000002/gb second · memory time
Compute available
unknown
Constrain compute
unknown
EU-only service
unknown
current
Verified 13 Jul 2026
Official price source
Cerebrium
Serverless
A100 (80GB)
A100
fixed · 1 GPU
Standard
scale to zero · standard · unspecified
additive components · not a complete comparable price
additive price not total
  • USD 0.000007/core second · cpu core time
  • USD 0.000583/second · gpu time
  • USD 0.000002/gb second · memory time
Compute available
unknown
Constrain compute
unknown
EU-only service
unknown
current
Verified 13 Jul 2026
Official price source
Cerebrium
Serverless
L40s
L40S
fixed · 1 GPU
Standard
scale to zero · standard · unspecified
additive components · not a complete comparable price
additive price not total
  • USD 0.000007/core second · cpu core time
  • USD 0.000542/second · gpu time
  • USD 0.000002/gb second · memory time
Compute available
unknown
Constrain compute
unknown
EU-only service
unknown
current
Verified 13 Jul 2026
Official price source
Cerebrium
Serverless
T4
T4
fixed · 1 GPU
Standard
scale to zero · standard · unspecified
additive components · not a complete comparable price
additive price not total
  • USD 0.000007/core second · cpu core time
  • USD 0.000164/second · gpu time
  • USD 0.000002/gb second · memory time
Compute available
unknown
Constrain compute
unknown
EU-only service
unknown
current
Verified 13 Jul 2026
Official price source
Fireworks AI
On-Demand Deployments
B200 180 GB GPU
NVIDIA B200
fixed · 1 GPU
Always on
always on · standard · dedicated
inclusive total · complete comparable price
  • USD 10/hour · gpu time
Compute available
unknown
Constrain compute
unknown
EU-only service
unknown
current
Verified 13 Jul 2026
Official price source
Fireworks AI
On-Demand Deployments
B200 180 GB GPU
NVIDIA B200
fixed · 1 GPU
Scale to zero
scale to zero · standard · dedicated
inclusive total · complete comparable price
  • USD 10/hour · gpu time
Compute available
unknown
Constrain compute
unknown
EU-only service
unknown
current
Verified 13 Jul 2026
Official price source
Fireworks AI
On-Demand Deployments
H200 141 GB GPU
NVIDIA H200
fixed · 1 GPU
Always on
always on · standard · dedicated
inclusive total · complete comparable price
  • USD 7/hour · gpu time
Compute available
unknown
Constrain compute
unknown
EU-only service
unknown
current
Verified 13 Jul 2026
Official price source
Fireworks AI
On-Demand Deployments
H200 141 GB GPU
NVIDIA H200
fixed · 1 GPU
Scale to zero
scale to zero · standard · dedicated
inclusive total · complete comparable price
  • USD 7/hour · gpu time
Compute available
unknown
Constrain compute
unknown
EU-only service
unknown
current
Verified 13 Jul 2026
Official price source
Fireworks AI
On-Demand Deployments
B300 288 GB GPU
NVIDIA B300
fixed · 1 GPU
Always on
always on · standard · dedicated
inclusive total · complete comparable price
  • USD 12/hour · gpu time
Compute available
unknown
Constrain compute
unknown
EU-only service
unknown
current
Verified 13 Jul 2026
Official price source
Fireworks AI
On-Demand Deployments
B300 288 GB GPU
NVIDIA B300
fixed · 1 GPU
Scale to zero
scale to zero · standard · dedicated
inclusive total · complete comparable price
  • USD 12/hour · gpu time
Compute available
unknown
Constrain compute
unknown
EU-only service
unknown
current
Verified 13 Jul 2026
Official price source
Fireworks AI
On-Demand Deployments
H100 80 GB GPU
H100
fixed · 1 GPU
Always on
always on · standard · dedicated
inclusive total · complete comparable price
  • USD 7/hour · gpu time
Compute available
unknown
Constrain compute
unknown
EU-only service
unknown
current
Verified 13 Jul 2026
Official price source
Fireworks AI
On-Demand Deployments
H100 80 GB GPU
H100
fixed · 1 GPU
Scale to zero
scale to zero · standard · dedicated
inclusive total · complete comparable price
  • USD 7/hour · gpu time
Compute available
unknown
Constrain compute
unknown
EU-only service
unknown
current
Verified 13 Jul 2026
Official price source
Hugging Face
Inference Endpoints
AWS NVIDIA L4 x4
L4
fixed · 4 GPU
Always on
always on · standard · dedicated
inclusive total · complete comparable price
  • USD 3.8/hour · compute time
Compute available
unknown
Constrain compute
unknown
EU-only service
unknown
current
Verified 13 Jul 2026
Official price source
Hugging Face
Inference Endpoints
AWS NVIDIA L4 x4
L4
fixed · 4 GPU
Scale to zero
scale to zero · standard · dedicated
inclusive total · complete comparable price
  • USD 3.8/hour · compute time
Compute available
unknown
Constrain compute
unknown
EU-only service
unknown
current
Verified 13 Jul 2026
Official price source
Hugging Face
Inference Endpoints
AWS NVIDIA T4 x4
T4
fixed · 4 GPU
Always on
always on · standard · dedicated
inclusive total · complete comparable price
  • USD 3/hour · compute time
Compute available
unknown
Constrain compute
unknown
EU-only service
unknown
current
Verified 13 Jul 2026
Official price source
Hugging Face
Inference Endpoints
AWS NVIDIA T4 x4
T4
fixed · 4 GPU
Scale to zero
scale to zero · standard · dedicated
inclusive total · complete comparable price
  • USD 3/hour · compute time
Compute available
unknown
Constrain compute
unknown
EU-only service
unknown
current
Verified 13 Jul 2026
Official price source
Hugging Face
Inference Endpoints
GCP NVIDIA A100 x4
A100
fixed · 4 GPU
Always on
always on · standard · dedicated
inclusive total · complete comparable price
  • USD 14.4/hour · compute time
Compute available
unknown
Constrain compute
unknown
EU-only service
unknown
current
Verified 13 Jul 2026
Official price source
Hugging Face
Inference Endpoints
GCP NVIDIA A100 x4
A100
fixed · 4 GPU
Scale to zero
scale to zero · standard · dedicated
inclusive total · complete comparable price
  • USD 14.4/hour · compute time
Compute available
unknown
Constrain compute
unknown
EU-only service
unknown
current
Verified 13 Jul 2026
Official price source
Hugging Face
Inference Endpoints
AWS NVIDIA A100 x8
A100
fixed · 8 GPU
Always on
always on · standard · dedicated
inclusive total · complete comparable price
  • USD 20/hour · compute time
Compute available
unknown
Constrain compute
unknown
EU-only service
unknown
current
Verified 13 Jul 2026
Official price source
Hugging Face
Inference Endpoints
AWS NVIDIA A100 x8
A100
fixed · 8 GPU
Scale to zero
scale to zero · standard · dedicated
inclusive total · complete comparable price
  • USD 20/hour · compute time
Compute available
unknown
Constrain compute
unknown
EU-only service
unknown
current
Verified 13 Jul 2026
Official price source
Hugging Face
Inference Endpoints
AWS NVIDIA A100 x1
A100
fixed · 1 GPU
Always on
always on · standard · dedicated
inclusive total · complete comparable price
  • USD 2.5/hour · compute time
Compute available
unknown
Constrain compute
unknown
EU-only service
unknown
current
Verified 13 Jul 2026
Official price source
Hugging Face
Inference Endpoints
AWS NVIDIA A100 x1
A100
fixed · 1 GPU
Scale to zero
scale to zero · standard · dedicated
inclusive total · complete comparable price
  • USD 2.5/hour · compute time
Compute available
unknown
Constrain compute
unknown
EU-only service
unknown
current
Verified 13 Jul 2026
Official price source
Hugging Face
Inference Endpoints
GCP NVIDIA A100 x2
A100
fixed · 2 GPU
Always on
always on · standard · dedicated
inclusive total · complete comparable price
  • USD 7.2/hour · compute time
Compute available
unknown
Constrain compute
unknown
EU-only service
unknown
current
Verified 13 Jul 2026
Official price source
Hugging Face
Inference Endpoints
GCP NVIDIA A100 x2
A100
fixed · 2 GPU
Scale to zero
scale to zero · standard · dedicated
inclusive total · complete comparable price
  • USD 7.2/hour · compute time
Compute available
unknown
Constrain compute
unknown
EU-only service
unknown
current
Verified 13 Jul 2026
Official price source
Hugging Face
Inference Endpoints
AWS NVIDIA H200 x2
NVIDIA H200
fixed · 2 GPU
Always on
always on · standard · dedicated
inclusive total · complete comparable price
  • USD 10/hour · compute time
Compute available
unknown
Constrain compute
unknown
EU-only service
unknown
current
Verified 13 Jul 2026
Official price source
Hugging Face
Inference Endpoints
AWS NVIDIA H200 x2
NVIDIA H200
fixed · 2 GPU
Scale to zero
scale to zero · standard · dedicated
inclusive total · complete comparable price
  • USD 10/hour · compute time
Compute available
unknown
Constrain compute
unknown
EU-only service
unknown
current
Verified 13 Jul 2026
Official price source
Hugging Face
Inference Endpoints
AWS NVIDIA H200 x8
NVIDIA H200
fixed · 8 GPU
Always on
always on · standard · dedicated
inclusive total · complete comparable price
  • USD 40/hour · compute time
Compute available
unknown
Constrain compute
unknown
EU-only service
unknown
current
Verified 13 Jul 2026
Official price source
Hugging Face
Inference Endpoints
AWS NVIDIA H200 x8
NVIDIA H200
fixed · 8 GPU
Scale to zero
scale to zero · standard · dedicated
inclusive total · complete comparable price
  • USD 40/hour · compute time
Compute available
unknown
Constrain compute
unknown
EU-only service
unknown
current
Verified 13 Jul 2026
Official price source
Hugging Face
Inference Endpoints
AWS NVIDIA A100 x4
A100
fixed · 4 GPU
Always on
always on · standard · dedicated
inclusive total · complete comparable price
  • USD 10/hour · compute time
Compute available
unknown
Constrain compute
unknown
EU-only service
unknown
current
Verified 13 Jul 2026
Official price source
Hugging Face
Inference Endpoints
AWS NVIDIA A100 x4
A100
fixed · 4 GPU
Scale to zero
scale to zero · standard · dedicated
inclusive total · complete comparable price
  • USD 10/hour · compute time
Compute available
unknown
Constrain compute
unknown
EU-only service
unknown
current
Verified 13 Jul 2026
Official price source
Hugging Face
Inference Endpoints
AWS NVIDIA H200 x4
NVIDIA H200
fixed · 4 GPU
Always on
always on · standard · dedicated
inclusive total · complete comparable price
  • USD 20/hour · compute time
Compute available
unknown
Constrain compute
unknown
EU-only service
unknown
current
Verified 13 Jul 2026
Official price source
Hugging Face
Inference Endpoints
AWS NVIDIA H200 x4
NVIDIA H200
fixed · 4 GPU
Scale to zero
scale to zero · standard · dedicated
inclusive total · complete comparable price
  • USD 20/hour · compute time
Compute available
unknown
Constrain compute
unknown
EU-only service
unknown
current
Verified 13 Jul 2026
Official price source
Hugging Face
Inference Endpoints
GCP NVIDIA L4 x1
L4
fixed · 1 GPU
Always on
always on · standard · dedicated
inclusive total · complete comparable price
  • USD 0.7/hour · compute time
Compute available
unknown
Constrain compute
unknown
EU-only service
unknown
current
Verified 13 Jul 2026
Official price source
Hugging Face
Inference Endpoints
GCP NVIDIA L4 x1
L4
fixed · 1 GPU
Scale to zero
scale to zero · standard · dedicated
inclusive total · complete comparable price
  • USD 0.7/hour · compute time
Compute available
unknown
Constrain compute
unknown
EU-only service
unknown
current
Verified 13 Jul 2026
Official price source
Hugging Face
Inference Endpoints
GCP NVIDIA A100 x8
A100
fixed · 8 GPU
Always on
always on · standard · dedicated
inclusive total · complete comparable price
  • USD 28.8/hour · compute time
Compute available
unknown
Constrain compute
unknown
EU-only service
unknown
current
Verified 13 Jul 2026
Official price source
Hugging Face
Inference Endpoints
GCP NVIDIA A100 x8
A100
fixed · 8 GPU
Scale to zero
scale to zero · standard · dedicated
inclusive total · complete comparable price
  • USD 28.8/hour · compute time
Compute available
unknown
Constrain compute
unknown
EU-only service
unknown
current
Verified 13 Jul 2026
Official price source
Hugging Face
Inference Endpoints
AWS NVIDIA A10G x1
A10G
fixed · 1 GPU
Always on
always on · standard · dedicated
inclusive total · complete comparable price
  • USD 1/hour · compute time
Compute available
unknown
Constrain compute
unknown
EU-only service
unknown
current
Verified 13 Jul 2026
Official price source
Hugging Face
Inference Endpoints
AWS NVIDIA A10G x1
A10G
fixed · 1 GPU
Scale to zero
scale to zero · standard · dedicated
inclusive total · complete comparable price
  • USD 1/hour · compute time
Compute available
unknown
Constrain compute
unknown
EU-only service
unknown
current
Verified 13 Jul 2026
Official price source
Hugging Face
Inference Endpoints
GCP NVIDIA H100 x2
H100
fixed · 2 GPU
Always on
always on · standard · dedicated
inclusive total · complete comparable price
  • USD 20/hour · compute time
Compute available
unknown
Constrain compute
unknown
EU-only service
unknown
current
Verified 13 Jul 2026
Official price source
Hugging Face
Inference Endpoints
GCP NVIDIA H100 x2
H100
fixed · 2 GPU
Scale to zero
scale to zero · standard · dedicated
inclusive total · complete comparable price
  • USD 20/hour · compute time
Compute available
unknown
Constrain compute
unknown
EU-only service
unknown
current
Verified 13 Jul 2026
Official price source
Hugging Face
Inference Endpoints
AWS NVIDIA T4 x1
T4
fixed · 1 GPU
Always on
always on · standard · dedicated
inclusive total · complete comparable price
  • USD 0.5/hour · compute time
Compute available
unknown
Constrain compute
unknown
EU-only service
unknown
current
Verified 13 Jul 2026
Official price source
Hugging Face
Inference Endpoints
AWS NVIDIA T4 x1
T4
fixed · 1 GPU
Scale to zero
scale to zero · standard · dedicated
inclusive total · complete comparable price
  • USD 0.5/hour · compute time
Compute available
unknown
Constrain compute
unknown
EU-only service
unknown
current
Verified 13 Jul 2026
Official price source
Hugging Face
Inference Endpoints
AWS NVIDIA L40S x8
L40S
fixed · 8 GPU
Always on
always on · standard · dedicated
inclusive total · complete comparable price
  • USD 23.5/hour · compute time
Compute available
unknown
Constrain compute
unknown
EU-only service
unknown
current
Verified 13 Jul 2026
Official price source
Hugging Face
Inference Endpoints
AWS NVIDIA L40S x8
L40S
fixed · 8 GPU
Scale to zero
scale to zero · standard · dedicated
inclusive total · complete comparable price
  • USD 23.5/hour · compute time
Compute available
unknown
Constrain compute
unknown
EU-only service
unknown
current
Verified 13 Jul 2026
Official price source
Hugging Face
Inference Endpoints
GCP NVIDIA T4 x1
T4
fixed · 1 GPU
Always on
always on · standard · dedicated
inclusive total · complete comparable price
  • USD 0.5/hour · compute time
Compute available
unknown
Constrain compute
unknown
EU-only service
unknown
current
Verified 13 Jul 2026
Official price source
Hugging Face
Inference Endpoints
GCP NVIDIA T4 x1
T4
fixed · 1 GPU
Scale to zero
scale to zero · standard · dedicated
inclusive total · complete comparable price
  • USD 0.5/hour · compute time
Compute available
unknown
Constrain compute
unknown
EU-only service
unknown
current
Verified 13 Jul 2026
Official price source
Hugging Face
Inference Endpoints
GCP NVIDIA H100 x1
H100
fixed · 1 GPU
Always on
always on · standard · dedicated
inclusive total · complete comparable price
  • USD 10/hour · compute time
Compute available
unknown
Constrain compute
unknown
EU-only service
unknown
current
Verified 13 Jul 2026
Official price source
Hugging Face
Inference Endpoints
GCP NVIDIA H100 x1
H100
fixed · 1 GPU
Scale to zero
scale to zero · standard · dedicated
inclusive total · complete comparable price
  • USD 10/hour · compute time
Compute available
unknown
Constrain compute
unknown
EU-only service
unknown
current
Verified 13 Jul 2026
Official price source
Hugging Face
Inference Endpoints
AWS NVIDIA L40S x1
L40S
fixed · 1 GPU
Always on
always on · standard · dedicated
inclusive total · complete comparable price
  • USD 1.8/hour · compute time
Compute available
unknown
Constrain compute
unknown
EU-only service
unknown
current
Verified 13 Jul 2026
Official price source
Hugging Face
Inference Endpoints
AWS NVIDIA L40S x1
L40S
fixed · 1 GPU
Scale to zero
scale to zero · standard · dedicated
inclusive total · complete comparable price
  • USD 1.8/hour · compute time
Compute available
unknown
Constrain compute
unknown
EU-only service
unknown
current
Verified 13 Jul 2026
Official price source
Hugging Face
Inference Endpoints
AWS NVIDIA A100 x2
A100
fixed · 2 GPU
Always on
always on · standard · dedicated
inclusive total · complete comparable price
  • USD 5/hour · compute time
Compute available
unknown
Constrain compute
unknown
EU-only service
unknown
current
Verified 13 Jul 2026
Official price source
Hugging Face
Inference Endpoints
AWS NVIDIA A100 x2
A100
fixed · 2 GPU
Scale to zero
scale to zero · standard · dedicated
inclusive total · complete comparable price
  • USD 5/hour · compute time
Compute available
unknown
Constrain compute
unknown
EU-only service
unknown
current
Verified 13 Jul 2026
Official price source