Shared agentsCoworker Intelligence
Coworker Intelligence combines your enterprise graph with the right model for each task, so answers and actions are grounded in your company's knowledge and permissions.
- AI Intelligence & Models
//Auto Routing
The right modelfor every request.
Auto Routing sends each Coworker request to the model that best fits its task, quality bar, latency and cost, with fallbacks when a provider is slow or down.
Mix
Every task matched to a model
Quick lookups go to fast models; analysis and long documents go to stronger reasoning models.
Spend
Quality kept, spend reduced
Routing avoids paying top-model prices for work a smaller model does just as well.
Budget
Routing that respects budgets
Routing rules weigh each team's budget, so spend stays within the limits admins set.
//Overview
What is LLM auto routing?
LLM auto routing automatically sends each AI request to the model best suited to it, weighing task type, quality, latency and cost, instead of using one large model for everything.
DataObserve Coworker Auto Routing classifies each request, routes it to an approved model, falls back to another provider during outages and learns from feedback and evaluations over time.
//How routing works
Classify, route, fall back,and learn.
Classify
Classify the request
Each request is tagged by task type, length, sensitivity and the quality it needs.
//Routing you control
Admins set the rules; routing does the rest
Approved models only
Routing chooses only from models your admins approve.
Sensitive data rules
Sensitive requests can be pinned to private or self-hosted models.
Manual override
Users can pick a model when they need a specific one.
Provider fallback
Outages at one provider don't stop work.
Cost visibility
Spend by model, team and task, in one view.
//Explore more withDataObserve
More from Coworker · AI Intelligence & Models
Model Hub
Model Hub gives Coworker users and builders a governed choice of frontier, open and DataObserve models, approved by admins and available across search, assistant and agents.
- AI Intelligence & Models
AI Gateway
The Coworker AI Gateway gives every AI tool your company uses one governed path to models, with authentication, rate limits, spend caps and logging in one place.
- AI Intelligence & Models
Usage Controls
Usage Controls let admins decide who can use which Coworker features, models and data, set budgets by team, and see how AI is used across the company.
- AI Intelligence & Models
//Use cases
Auto Routing: use cases
Cost reduction
Stop paying top-model prices for simple lookups.
Faster answers
Serve quick questions from fast models.
Resilience
Keep working through a provider outage.
Sensitive data
Pin regulated requests to private or self-hosted models.
//FAQs
Auto Routing: questions teams ask
Does auto routing lower quality?
No. Routing picks the cheapest model that meets the quality bar for the task, and quality is evaluated continuously.
Can users choose a model manually?
Yes. Users can override routing when they need a specific model.
What happens if a model provider goes down?
Requests fall back to the next approved model automatically.
Who decides which models routing can use?
Admins. Routing only chooses from models approved in Model Hub.
