Provider pricing

Inferless AI API pricing

Inferless is a serverless GPU platform for deploying custom LLMs, embedding models, and computer-vision workloads with per-second billing. Specializes in fast cold starts and autoscaling to zero.

Founded2023
HeadquartersSan Francisco, CA
Tracked modelsDirectory

Best for

Teams hosting custom or fine-tuned models without managing GPU infra.

Strengths

  • Per-second GPU billing with scale-to-zero
  • Sub-10 second cold starts on large models
  • Bring-your-own model from Hugging Face or container
  • Built-in monitoring, A/B routing, and webhooks