🔀

🔧 Developer Tools

Multi-Model Router

Route AI requests to the best model per task

AI LLM Pipeline Persistent Rule Storage Multi-Model Support Side-by-Side Compare Fallback Chains

About

Intelligent model routing based on task type. Configure rules to automatically select the best model for each request — send code questions to the 70B model, creative tasks to a high-temperature 8B, and everything else to a balanced default.

Compare responses across models side-by-side, set fallback chains, and optimize cost vs. quality per task type.

API Endpoints

POST /chat Send a chat request (auto-routed to best model)

POST /compare Compare responses across multiple models

POST /rules Configure routing rules

GET /models List available models

GET /health Health check

How It Works

Task Classification

Incoming request classified by task_type (code, creative, analysis, general).

Rule Matching

Routing rules loaded from persistent storage — task type mapped to target model.

Model Dispatch

Request forwarded to the selected AI model.

Fallback Handling

If primary model fails, request automatically retried with the fallback model.

Use Cases

💰

Cost Optimization

Route simple queries to smaller models and complex ones to larger models automatically.

🔬

A/B Testing

Compare model outputs side-by-side to evaluate quality for your use case.

🔗

Fallback Chains

Set primary and fallback models so requests never fail due to model unavailability.

🎯

Task Specialization

Route code, creative writing, and analysis to purpose-optimized models.

Quick Launch arrow_forward

Opens Aerostack dashboard to deploy this template

What's Included

check AI LLM Pipeline

check Persistent Rule Storage

check Multi-Model Support

check Side-by-Side Compare

check Fallback Chains

check 5 API endpoints

check Edge deployed

Pipeline

psychology LLM — AI text generation

Billing Model

metered

Pay per token used. Free tier included.