- Task
- Requested
- Routed to
- Tokens
- Quality
Example request. Costs use list prices for the token counts shown; certainty and quality are illustrative.
Rouva routes each request to the right model for the task, and lets owners set which models each teammate can use and how much they can spend.
Example request. Costs use list prices for the token counts shown; certainty and quality are illustrative.
Assign each member their models and a monthly spend cap. Rouva enforces both on every request.
Swap your provider SDK for Rouva's. With Intelligent Routing on, name a model and Rouva routes to a cheaper one when the task allows, or leave the model out and let it choose. Building an agent? Rouva can run the whole tool loop for you.
import OpenAI from 'openai' const openai = new OpenAI({ apiKey: process.env.OPENAI_API_KEY }) const res = await openai.chat.completions.create({ messages, model: 'gpt-4o', })
// npm install @rouvanpm/rouva import { Rouva } from '@rouvanpm/rouva' const rouva = new Rouva({ apiKey: process.env.ROUVA_API_KEY }) const res = await rouva.chat.completions.create({ messages, model: 'gpt-4o', // optional: omit and Rouva picks })
The right model for the task, picked per request.
Spend caps per key and member, budgets per agent run.
Quality scores and a plain-language routing reason.
Providers bill you directly; Rouva never marks up tokens. Keys are encrypted at rest, and conversation content is held only in short-lived caches.