Skip to content

Compute

Run any pricing function. Skip the infra.

Compute is the managed runtime that executes your pricing functions — rule engines, ML models, LLM prompts, external APIs, or composites. Model Profiling captures latency, throughput, and cost for every call so you always know which function to route to.

last-price.ai/compute
Compute fleet view — function, runtime, p95, error rate, cost.

What you can do

  • Deploy custom pricing functions with the same execution_type contract as built-ins.
  • Get GPU runtimes for ML pricing models — no machine to provision.
  • Profile every model: latency p50/p95/p99, throughput, error rate, and cost per call.
  • Promote and rollback model versions without touching client code.
  • Wire profiling output straight into Routing Policies — cheapest, fastest, weighted.

Ready to try it?

Spin up a free playground tenant in under a minute. No credit card.