Understand the mechanism

Define task cases and hard gates; score candidates on quality, latency, cost, context needs, tool/schema support, and vendor constraints; keep a fallback model.

Read the model selection path as a pipeline for the invoice-extraction service. At each stage, name the representation, the owner, and how information can be lost. Identifiers must mark prompt versions, model versions, indexes, and policies so “randomness” is not the default explanation for every bug in model selection.

If model L hits 96% field-F1 at $0.12/1k docs and model H hits 97% at $0.55/1k, the +1 pt may lose unless errors are extremely costly. Scoped specifically to model selection / invoice-extraction service / mechanism.

Keep the unit and the denominator visible when you discuss model selection. A percentage without a base, or a latency without a percentile, hides the failure mode this chapter cares about on the invoice-extraction service.

Learned stages estimate; deterministic stages enforce. A fluent result from the invoice-extraction service does not prove model selection used the right evidence. Preserve intermediates when privacy allows—candidate lists, traces, scores, citations—so you can see the first broken stage in the model selection path.

Operational correctness for model selection includes deadlines on the invoice-extraction service. If stages that feed this workload sum past the user’s patience, trim earlier—usually pack less, retrieve less, or parallelize—before blaming the model vendor for model selection. Mechanism diagrams that ignore time are incomplete.

Also pin one numeric memory from this model selection chapter: If model L hits 96% field-F1 at $0.12/1k docs and model H hits 97% at $0.55/1k, the +1 pt may lose unless errors are extremely costly. That number is not decoration; it is a template for how claims about model selection on the invoice-extraction service should look in design docs. Scoped specifically to model selection / invoice-extraction service / mechanism.

People confuse model selection with neighboring buzzwords when debugging the invoice-extraction service. Before changing prompts, ask whether the broken stage was evidence gathering, the model selection judgment itself, validation, or the product action. Fixing the wrong stage creates folklore (“we tried model selection and it failed”) that blocks the next team on the invoice-extraction service. Scoped specifically to model selection / invoice-extraction service / mechanism.

Write a five-line artifact for this page: goal, inputs, check, owner, stop rule. Invent one fluent failure that the check would catch. Keep details specific to choosing a model rather than generic AI advice.