# Model Inference / Latency/Cost

> "model-inference" (correctness-core layer) is traded against "Latency/Cost" (performance-core layer) — a principle cannot be scope-separated from a quality, metric, or cost it competes with; resolve by measuring "Latency/Cost" and choosing an explicit operating point.

Record: `tension:latency-cost-model-inference`
Kind: tension
Canonical: https://banes-lab.com/ontology/schema#tension-latency-cost-model-inference

## tensions-with

- [Model Inference](https://banes-lab.com/records/arch/model-inference.md)
- [Latency/Cost](https://banes-lab.com/records/lex/latency-cost.md)