Skip to content

27 offerings from 8 providers across serverless products and GPU instances.

39 complete comparable prices; incomplete and additive prices remain visible but are not ranked as totals.

8 of 8 serverless offerings include a scale-to-zero operating mode.

Published configurations: 24 GB

Serverless prices

All serverless GPUs →
Baseten.co
L4 · 24 GB VRAM/GPU
L4 · Serverless · fixed
  • USD 0.01414/minute
    compute time
Component price — not a total
L4 standard
scale to zero · dedicated
1 minute increments · bills deploying, scaling up, scaling down, making predictions
Cerebrium
1× L4 · VRAM unknown
L4 · Serverless · fixed
  • USD 0.00000655/core second
    cpu core time
  • USD 0.000222/second
    gpu time
  • USD 0.00000222/gb second
    memory time
Component price — not a total
Standard scale-to-zero
scale to zero
1 second increments
Enterprise pricing is listed as Custom and was omitted.
1 geography evidence item
Cerebrium
1× L4 · VRAM unknown
L4 · Serverless · fixed
  • USD 0.00000655/core second
    cpu core time
  • USD 0.000222/second
    gpu time
  • USD 0.00000222/gb second
    memory time
Component price — not a total
Standard
scale to zero
1 second increments · bills active compute time
EU compute EU-constrained
Sweden
Enterprise pricing is listed as Custom and was omitted.
3 geography evidence items
Google Cloud
1× L4 · 24 GB VRAM/GPU
NVIDIA-L4 · Serverless · fixed
  • USD 0.000018/core second
    cpu core time
  • USD 0.0001867/second
    gpu time
  • USD 0.000002/gib second
    memory time
Component price — not a total
NVIDIA-L4 No zonal redundancy
scale to zero
100 ms increments · 1 minute minimum · bills entire instance lifecycle
stale Verified 17 Sept 2026 Official price source (opens in new tab)
1 geography evidence item
Google Cloud
1× L4 · 24 GB VRAM/GPU
NVIDIA-L4 · Serverless · fixed
  • USD 0.000018/core second
    cpu core time
  • USD 0.0002909/second
    gpu time
  • USD 0.000002/gib second
    memory time
Component price — not a total
NVIDIA-L4 Zonal redundancy
scale to zero
100 ms increments · 1 minute minimum · bills entire instance lifecycle
stale Verified 17 Sept 2026 Official price source (opens in new tab)
1 geography evidence item
Google Cloud
1× L4 · 24 GB VRAM/GPU
NVIDIA-L4 · Serverless · fixed
  • USD 0.000018/core second
    cpu core time
  • USD 0.0001867/second
    gpu time
  • USD 0.000002/gib second
    memory time
Component price — not a total
NVIDIA-L4 No zonal redundancy
scale to zero
100 ms increments · 1 minute minimum · bills entire instance lifecycle
EU compute EU-constrained
Belgium · Netherlands
stale Verified 17 Sept 2026 Official price source (opens in new tab)
5 geography evidence items
Google Cloud
1× L4 · 24 GB VRAM/GPU
NVIDIA-L4 · Serverless · fixed
  • USD 0.000018/core second
    cpu core time
  • USD 0.0002909/second
    gpu time
  • USD 0.000002/gib second
    memory time
Component price — not a total
NVIDIA-L4 Zonal redundancy
scale to zero
100 ms increments · 1 minute minimum · bills entire instance lifecycle
EU compute EU-constrained
Belgium · Netherlands
stale Verified 17 Sept 2026 Official price source (opens in new tab)
5 geography evidence items
Modal
1× L4 · VRAM unknown
Nvidia L4 · Serverless · fixed
  • USD 0.0000131/core second → USD 0.00001507/core second effective
    cpu core time
  • USD 0.000222/second → USD 0.0002553/second effective
    gpu time
  • USD 0.00000222/gib second → USD 0.000002553/gib second effective
    memory time
Component price — not a total
Broad region
scale to zero
none minimum · bills running
stale Verified 17 Sept 2026 Official price source (opens in new tab)
5 geography evidence items
Modal
1× L4 · VRAM unknown
Nvidia L4 · Serverless · fixed
  • USD 0.0000131/core second → USD 0.00002293/core second effective
    cpu core time
  • USD 0.000222/second → USD 0.0003885/second effective
    gpu time
  • USD 0.00000222/gib second → USD 0.000003885/gib second effective
    memory time
Component price — not a total
Narrow region
scale to zero
none minimum · bills running
stale Verified 17 Sept 2026 Official price source (opens in new tab)
5 geography evidence items
Modal
1× L4 · VRAM unknown
Nvidia L4 · Serverless · fixed
  • USD 0.0000131/core second
    cpu core time
  • USD 0.000222/second
    gpu time
  • USD 0.00000222/gib second
    memory time
Component price — not a total
Standard
scale to zero
none minimum · bills running
stale Verified 17 Sept 2026 Official price source (opens in new tab)
5 geography evidence items
OVHcloud
1× L4 · 24 GB VRAM/GPU
l4-1-gpu · Serverless · fixed
  • USD 0.91/hour
    compute time
≈ EUR 0.8014/hour
scale-to-zero
scale to zero · dedicated
1 minute increments · bills SCALING, RUNNING

Instance prices

All GPU instances →
Google Cloud
g2-standard-4
1× L4
24 GB VRAM/GPU · 4 vCPUs · 16 GB RAM · storage unknown
Region: Council Bluffs, Iowa · US · us-central1
  • USD 0.7068/hour
    compute time
≈ EUR 0.6225/hour
US-CENTRAL1 on-demand
standard
Google Cloud
g2-standard-4
1× L4
24 GB VRAM/GPU · 4 vCPUs · 16 GB RAM · storage unknown
Region: Eemshaven, Netherlands · NL · europe-west4 · EU
  • USD 0.7429/hour
    compute time
≈ EUR 0.6543/hour
EUROPE-WEST4 on-demand
standard
Google Cloud
g2-standard-8
1× L4
24 GB VRAM/GPU · 8 vCPUs · 32 GB RAM · storage unknown
Region: Council Bluffs, Iowa · US · us-central1
  • USD 0.8536/hour
    compute time
≈ EUR 0.7518/hour
US-CENTRAL1 on-demand
standard
Google Cloud
g2-standard-8
1× L4
24 GB VRAM/GPU · 8 vCPUs · 32 GB RAM · storage unknown
Region: Eemshaven, Netherlands · NL · europe-west4 · EU
  • USD 0.8972/hour
    compute time
≈ EUR 0.7901/hour
EUROPE-WEST4 on-demand
standard
Google Cloud
g2-standard-12
1× L4
24 GB VRAM/GPU · 12 vCPUs · 48 GB RAM · storage unknown
Region: Council Bluffs, Iowa · US · us-central1
  • USD 1/hour
    compute time
≈ EUR 0.881/hour
US-CENTRAL1 on-demand
standard
Google Cloud
g2-standard-12
1× L4
24 GB VRAM/GPU · 12 vCPUs · 48 GB RAM · storage unknown
Region: Eemshaven, Netherlands · NL · europe-west4 · EU
  • USD 1.051/hour
    compute time
≈ EUR 0.926/hour
EUROPE-WEST4 on-demand
standard
Google Cloud
g2-standard-16
1× L4
24 GB VRAM/GPU · 16 vCPUs · 64 GB RAM · storage unknown
Region: Council Bluffs, Iowa · US · us-central1
  • USD 1.147/hour
    compute time
≈ EUR 1.01/hour
US-CENTRAL1 on-demand
standard
Google Cloud
g2-standard-16
1× L4
24 GB VRAM/GPU · 16 vCPUs · 64 GB RAM · storage unknown
Region: Eemshaven, Netherlands · NL · europe-west4 · EU
  • USD 1.206/hour
    compute time
≈ EUR 1.062/hour
EUROPE-WEST4 on-demand
standard
Google Cloud
g2-standard-32
1× L4
24 GB VRAM/GPU · 32 vCPUs · 128 GB RAM · storage unknown
Region: Council Bluffs, Iowa · US · us-central1
  • USD 1.734/hour
    compute time
≈ EUR 1.527/hour
US-CENTRAL1 on-demand
standard
Google Cloud
g2-standard-32
1× L4
24 GB VRAM/GPU · 32 vCPUs · 128 GB RAM · storage unknown
Region: Eemshaven, Netherlands · NL · europe-west4 · EU
  • USD 1.823/hour
    compute time
≈ EUR 1.605/hour
EUROPE-WEST4 on-demand
standard
Google Cloud
g2-standard-24
2× L4
24 GB VRAM/GPU · 24 vCPUs · 96 GB RAM · storage unknown
Region: Council Bluffs, Iowa · US · us-central1
  • USD 2.001/hour
    compute time
≈ EUR 1.762/hour
US-CENTRAL1 on-demand
standard
Google Cloud
g2-standard-24
2× L4
24 GB VRAM/GPU · 24 vCPUs · 96 GB RAM · storage unknown
Region: Eemshaven, Netherlands · NL · europe-west4 · EU
  • USD 2.103/hour
    compute time
≈ EUR 1.852/hour
EUROPE-WEST4 on-demand
standard
Google Cloud
g2-standard-48
4× L4
24 GB VRAM/GPU · 48 vCPUs · 192 GB RAM · storage unknown
Region: Council Bluffs, Iowa · US · us-central1
  • USD 4.002/hour
    compute time
≈ EUR 3.524/hour
US-CENTRAL1 on-demand
standard
Google Cloud
g2-standard-48
4× L4
24 GB VRAM/GPU · 48 vCPUs · 192 GB RAM · storage unknown
Region: Eemshaven, Netherlands · NL · europe-west4 · EU
  • USD 4.206/hour
    compute time
≈ EUR 3.704/hour
EUROPE-WEST4 on-demand
standard
Google Cloud
g2-standard-96
8× L4
24 GB VRAM/GPU · 96 vCPUs · 384 GB RAM · storage unknown
Region: Council Bluffs, Iowa · US · us-central1
  • USD 8.003/hour
    compute time
≈ EUR 7.048/hour
US-CENTRAL1 on-demand
standard
Google Cloud
g2-standard-96
8× L4
24 GB VRAM/GPU · 96 vCPUs · 384 GB RAM · storage unknown
Region: Eemshaven, Netherlands · NL · europe-west4 · EU
  • USD 8.412/hour
    compute time
≈ EUR 7.408/hour
EUROPE-WEST4 on-demand
standard

Available Providers