RunPod serverless and GPU pricing

Where this provider fits

A GPU cloud with pods, serverless workers, endpoint controls, and community templates for custom AI workloads.

What can change the bill

• Total cost depends on startup, idle, and execution time
• It is not directly comparable with per-output hosted APIs

Who should evaluate this route

• Serverless and persistent GPU options
• Control over model packaging
• Broad GPU selection

Compare two nearby options

Use the same model, meter, configuration, and workload when comparing channels.

• fal.ai pricing evidenceA developer platform focused on fast inference for image, video, and audio models, with model-level pricing and serverless APIs.
• Replicate pricing evidenceA model marketplace and deployment platform that exposes public and custom models through a consistent prediction API.

Recorded price evidence

Checked evidence, not a ranking: fresh rows were checked against the linked first-party source on the shown date. APIDir does not name a cheapest provider unless exact configurations and units match. Recheck before purchasing.

Model price comparisons for this provider

Follow each model page to compare this provider route with other accepted, source-linked offers on matching billing meters.

• GPT-OSS 120B provider pricingCompare source-linked GPT-OSS 120B input and output token prices across provider routes while keeping infrastructure compute meters separate.

Related provider comparisons

Use a named alternative and explicit decision dimensions instead of comparing provider brands in the abstract.

• RunPod vs ReplicateCompare abstraction and control before choosing a route.