GPU Router
Loading models…

Fetching the model catalog.

The home for
excess compute.

The models you want. At the prices you should pay.
Find capacity, track prices, and route through one API.

THE OPEN MARKET FOR INTELLIGENCE

Find compatible models and migrate your code.

THE SAME INTELLIGENCE. LESS SPEND.

Model pricing

PAST 24 HOURS
Publisher direct
OpenRouter catalog
Our price

1M input + 1M output tokens · USD

Loading recorded prices…
Awaiting a price

Connecting to model catalogs and market prices…

Models to discover
Models with live discounts
Market-reported spend*
Checking routing availability
WHERE DEMAND GOES

Models in demand.

reported across 0 tracked models
No fresh model activity is available.

Source: market activity · current reported counters, labeled 24h upstream. A rolling window has not been confirmed. Top 0 models shown; the share bar and totals include all 0.

MORE MODEL FOR YOUR MONEY

The biggest savings.

View rankings ↗

Current discounts from direct list prices.

0% off50%100%
No fresh discounts are available.

1M input + 1M output tokens. Discounts use observed market prices and matching direct references; stale quotes are excluded.

FIND YOUR NEXT MODEL

Discover more. Spend less.

Browse all models ↗

Models with current discounts will appear here when fresh prices are available.

SAVINGS, OVER TIME

See how much you save.

Compare market discounts against direct list prices. Higher means more savings.

AWAITING DATA
Discount vs direct list price · higher is better
Loading recorded prices…

Higher is better. 99% off means paying 1% of direct list price. Each discount uses the direct price recorded at that time; negative values mean above-list pricing. Recorded observations only; gaps stay visible. In + out = 1M input + 1M output tokens.

Biggest discounts

Market prices against direct list prices.

View all ↗
ModelOff direct

Waiting for qualified market observations.

Intelligence per dollar

GPQA Diamond score ÷ in + out price.

View all ↗
ModelScore / $

Waiting for qualified market observations.

Models in demand

Market-reported request activity.*

View all ↗
ModelRequests*

Waiting for qualified market observations.

Intelligence leaders

Published GPQA Diamond observations.

View all ↗
ModelGPQA score

Waiting for qualified market observations.

BUILT FOR BUILDERS

A new base URL.
A better price.

Keep your OpenAI-compatible client. Choose an exact model ID, connect a key, and let GPU Router find the eligible route.

Read the quickstart →
// Your next request
const completion = await client.chat.completions.create({
 model: "openai/gpt-6-astra",
 messages: [{ role: "user",
 content: "Build something great." }]
});

* The market feed labels activity as 24h; the rolling window has not been confirmed. Other venues do not expose comparable public model-wide usage, so totals are not combined. Benchmark observations: Sep 6, 2026; external endpoints, not marketplace quality guarantees.