DeployWhere

Replicate

API platform for running, fine-tuning, and deploying machine learning models.

Replicate is a GPU and AI hosting provider based in United States. No European datacenter locations are recorded for it yet. There is no free tier. As a US company it falls under the US CLOUD Act. Pricing on this page is published only after a human has reviewed it.

Company based in
United States
European regions
None listed
US CLOUD Act
US company
GDPR agreement
Available
Free tier
None
VAT
Shown excluding VAT
Replicate website
Website image published by Replicate.

Pricing

28 plans
  • anthropic/claude-3.7-sonnet
    Monthly
    $0.015/ thousand output tokens; $3.00/ million input tokens
    vCPU
    Not listed
    Memory
    Not listed
    Storage
    Not listed
  • black-forest-labs/flux-1.1-pro
    Monthly
    $0.04/ output image
    vCPU
    Not listed
    Memory
    Not listed
    Storage
    Not listed
  • black-forest-labs/flux-dev
    Monthly
    $0.025/ output image
    vCPU
    Not listed
    Memory
    Not listed
    Storage
    Not listed
  • black-forest-labs/flux-schnell
    Monthly
    $3.00/ thousand output images
    vCPU
    Not listed
    Memory
    Not listed
    Storage
    Not listed
  • deepseek-ai/deepseek-r1
    Monthly
    $0.01/ thousand output tokens; $3.75/ million input tokens
    vCPU
    Not listed
    Memory
    Not listed
    Storage
    Not listed
  • ideogram-ai/ideogram-v3-quality
    Monthly
    $0.09/ output image
    vCPU
    Not listed
    Memory
    Not listed
    Storage
    Not listed
  • recraft-ai/recraft-v3
    Monthly
    $0.04/ output image
    vCPU
    Not listed
    Memory
    Not listed
    Storage
    Not listed
  • wavespeedai/wan-2.1-i2v-480p
    Monthly
    $0.09/ second of output video
    vCPU
    Not listed
    Memory
    Not listed
    Storage
    Not listed
  • wavespeedai/wan-2.1-i2v-720p
    Monthly
    $0.25/ second of output video
    vCPU
    Not listed
    Memory
    Not listed
    Storage
    Not listed
  • CPU (Small)
    Also priced at $0.000025/sec.
    Monthly
    about 65.7 USD$0.09/hr
    vCPU
    1
    Memory
    2 GB
    Storage
    Not listed
  • CPU
    Also priced at $0.000100/sec.
    Monthly
    about 262.8 USD$0.36/hr
    vCPU
    4
    Memory
    8 GB
    Storage
    Not listed
  • Nvidia A100 (80GB) GPU
    Also priced at $0.001400/sec. GPU RAM is 80GB.
    Monthly
    about 3679.2 USD$5.04/hr
    vCPU
    10
    Memory
    144 GB
    Storage
    Not listed
  • 2x Nvidia A100 (80GB) GPU
    Also priced at $0.002800/sec. GPU RAM is 160GB.
    Monthly
    about 7358.4 USD$10.08/hr
    vCPU
    20
    Memory
    288 GB
    Storage
    Not listed
  • Nvidia H100 GPU
    Also priced at $0.001525/sec. GPU RAM is 80GB.
    Monthly
    about 4007.7 USD$5.49/hr
    vCPU
    13
    Memory
    144 GB
    Storage
    Not listed
  • Nvidia L40S GPU
    Also priced at $0.000975/sec. GPU RAM is 48GB.
    Monthly
    about 2562.3 USD$3.51/hr
    vCPU
    10
    Memory
    65 GB
    Storage
    Not listed
  • 2x Nvidia L40S GPU
    Also priced at $0.001950/sec. GPU RAM is 96GB.
    Monthly
    about 5124.6 USD$7.02/hr
    vCPU
    20
    Memory
    144 GB
    Storage
    Not listed
  • Nvidia T4 GPU
    Also priced at $0.000225/sec. GPU RAM is 16GB.
    Monthly
    about 591.3 USD$0.81/hr
    vCPU
    4
    Memory
    16 GB
    Storage
    Not listed
  • 4x Nvidia A100 (80GB) GPU
    Also priced at $0.005600/sec. Additional Multi-GPU A100 capacity is available with committed spend contracts.
    Monthly
    about 14716.8 USD$20.16/hr
    vCPU
    Not listed
    Memory
    Not listed
    Storage
    Not listed
  • 8x Nvidia A100 (80GB) GPU
    Also priced at $0.011200/sec. Additional Multi-GPU A100 capacity is available with committed spend contracts.
    Monthly
    about 29433.6 USD$40.32/hr
    vCPU
    Not listed
    Memory
    Not listed
    Storage
    Not listed
  • 2x Nvidia H100 GPU
    Also priced at $0.003050/sec. Additional Multi-GPU H100 capacity is available with committed spend contracts.
    Monthly
    about 8015.4 USD$10.98/hr
    vCPU
    Not listed
    Memory
    Not listed
    Storage
    Not listed
  • 4x Nvidia H100 GPU
    Also priced at $0.006100/sec. Additional Multi-GPU H100 capacity is available with committed spend contracts.
    Monthly
    about 16030.8 USD$21.96/hr
    vCPU
    Not listed
    Memory
    Not listed
    Storage
    Not listed
  • 8x Nvidia H100 GPU
    Also priced at $0.012200/sec. Additional Multi-GPU H100 capacity is available with committed spend contracts.
    Monthly
    about 32061.6 USD$43.92/hr
    vCPU
    Not listed
    Memory
    Not listed
    Storage
    Not listed
  • Nvidia H200 GPU
    Also priced at $0.001525/sec. H200 capacity is available with committed spend contracts.
    Monthly
    about 4007.7 USD$5.49/hr
    vCPU
    Not listed
    Memory
    Not listed
    Storage
    Not listed
  • 2x Nvidia H200 GPU
    Also priced at $0.003050/sec. Additional Multi-GPU H200 capacity is available with committed spend contracts.
    Monthly
    about 8015.4 USD$10.98/hr
    vCPU
    Not listed
    Memory
    Not listed
    Storage
    Not listed
  • 4x Nvidia H200 GPU
    Also priced at $0.006100/sec. Additional Multi-GPU H200 capacity is available with committed spend contracts.
    Monthly
    about 16030.8 USD$21.96/hr
    vCPU
    Not listed
    Memory
    Not listed
    Storage
    Not listed
  • 8x Nvidia H200 GPU
    Also priced at $0.012200/sec. Additional Multi-GPU H200 capacity is available with committed spend contracts.
    Monthly
    about 32061.6 USD$43.92/hr
    vCPU
    Not listed
    Memory
    Not listed
    Storage
    Not listed
  • 4x Nvidia L40S GPU
    Also priced at $0.003900/sec. Additional Multi-GPU L40S capacity is available with committed spend contracts.
    Monthly
    about 10249.2 USD$14.04/hr
    vCPU
    Not listed
    Memory
    Not listed
    Storage
    Not listed
  • 8x Nvidia L40S GPU
    Also priced at $0.007800/sec. Additional Multi-GPU L40S capacity is available with committed spend contracts.
    Monthly
    about 20498.4 USD$28.08/hr
    vCPU
    Not listed
    Memory
    Not listed
    Storage
    Not listed

Pros and cons

What works

  • Usage-based billing with automatic scaling to zero for custom deployments.
  • Public models and custom model deployment are available through one API.
  • Enterprise plans offer dedicated support, higher GPU limits, volume discounts, and performance SLAs.

What to watch

  • No recurring monthly plans are published.
  • Costs vary by model, hardware, runtime, or input and output volume.
  • Enterprise pricing is not published.
  • Datacenter locations and uptime figures are not stated.

Specifications

Supported runtimes
Node, Python, HTTP
Support
Standard support details are not stated. Enterprise plans offer dedicated support, priority support, a dedicated account manager, and performance SLAs.

About Replicate

Replicate provides API access to public models and deployment of custom models using Cog. Public models are billed by hardware time or model inputs and outputs, while private models generally use dedicated hardware billed for online time. The source pages do not state the operator's domicile, datacenter locations, or data residency guarantees.

Reviews

None yet

No approved reviews yet. Every review is read by a person before it appears, and only approved reviews count toward a rating.

Other gpu and ai hosting

Something here out of date?

Every figure on this page was checked by a person, and providers change their plans without telling anybody. Say what moved and it gets checked again.

Work at Replicate? Claim this listing to correct the description and add screenshots. Verification is a confirmation link to an address on your own domain. Pricing still goes through the same check as every other listing here.