Skip to content
StackShip

Replicate

Replicate is a cloud platform for running and fine-tuning open-source and proprietary machine learning models through an API, billed by usage.

Verified
Visit Replicate ↗

Our verdict

Run thousands of open-source and proprietary models through an API, billed by run time or per output. Ideal when you need a specific open model or fine-tuning.

Best for

  • ✓ Running open-source models via API
  • ✓ Fine-tuning

Watch out for

  • ! Private models bill for all uptime, including idle time

Pricing

Read from the official pricing page. Prices in USD; always confirm on the vendor's site.

PlanPriceDetails
Pay as you go$0 + usagePublic models are billed by run time on their hardware (e.g. Nvidia T4 $0.000225/sec = $0.81/hr; H100 $0.001525/sec = $5.49/hr) or per input/output unit (e.g. flux-dev $0.025/image). Private models are billed for all uptime, including setup and idle time
EnterpriseCustomVolume discounts on request. Check site.

Note: The pricing page does not mention a free tier or free credits (check site).

Featured in

Other image & video ai