AI Router — Account

Balance

—

Charged per request. No subscriptions.

💡

Estimated Cost Savings

Smart Routing

$0

0% savings vs. naive baseline

Est. Monthly Savings

$0

vs. naive baseline

How we estimate: We compare sending every request to the most expensive model in your enabled pool vs. InfraX Router's task-aware routing (which picks the cheapest model that meets quality thresholds per task type: chat → Llama 3.1 8B, coding → DeepSeek, reasoning → DeepSeek R1, long context → Gemini Flash, etc.). Based on typical task mix: chat 35%, coding 20%, general 20%, reasoning 15%, math 10%.

How we calculate →

How we estimate: Based on your enabled models and typical task distribution. Real savings depend on your actual task mix, token usage, cache hits, and quality thresholds. Cache hits = $0. Fallbacks to premium models may reduce savings. Learn more.

Add credits

💰

Real Savings vs. Naive Baseline

Live Data

$0.00/1M tokens

0% savings vs. naive baseline

Est. Monthly Savings

$0.00

vs. naive baseline

Real-time estimate: Based on your enabled models and current task distribution (chat 35%, coding 20%, general 20%, reasoning 15%, math 10%). Compares naive baseline (always most expensive model) vs. InfraX task-aware routing.

How we calculate →

Real-time data: From /billing/savings/public. Cache hits = $0. Fallbacks to premium models may reduce savings. Learn more.

🔑 API key

Model preferences

OpenRouter only · Free models excluded

Choose which models are available for routing. When empty, all models are eligible. Select specific models to restrict routing to only those.

Balance

Estimated Cost Savings

Add credits

Real Savings vs. Naive Baseline

🔑 API key

Model preferences

Estimated $0/mo savings

🎯 Recommended for you

All available models

Recent usage