> ## Documentation Index
> Fetch the complete documentation index at: https://docs.bourneapp.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Provider and model selection

> Choose the right provider-model pair for quality, cost, and latency tradeoffs.

Choose your provider-model pair by workload, not by raw benchmark claims.

## Decision framework

* Quality-critical workflows: pick higher-capability models.
* Cost-sensitive automation: pick efficient mid-tier models.
* Low-latency loops: pick fast variants.

## Start safe, then optimize

Start with one stable model, then tune:

1. Response quality
2. Latency
3. Cost per task

## Keep migration simple

When switching models, change one variable at a time and observe effects before further changes.