Cloud Platform

Features

Data usePut your data to use

Predict future trends using historical data. Route messages, detect spam and recommend products for users by using Prediction API. Google’s powerful infrastructure analyzes the data for you.

FastFast and reliable

Your data is replicated across multiple data centers using Cloud Storage. Most prediction queries take less than 200ms. Greater performance is available. See the documentation for performance tips.

Cloud integrationCloud integration

Google services are designed to work together. Use results from Google BigQuery or pull in datasets of up to 2.5GB from Cloud Storage to create a predictive model. Use Prediction API directly from Google App Engine.

Dev toolsPowerful development tools

Prediction API can integrate with App Engine, and the RESTful API is available through libraries for many popular languages, such as Python, JavaScript and .NET. See the documentation for the full list.

SupportExamples and support

Learn more about Prediction API by checking out our example applications. Take a look at our demos and featured applications for more ideas.

PricingFlexible pricing

Limited usage is free for the first six months. Beyond that, Prediction API’s comprehensible pricing comes with a 99.9% availability service level agreement. Like other Cloud products, you pay for what you use.

Pricing calculator

The Pricing Calculator provides you with a simple tool that can help you get a sense of what an application running on Google Cloud Platform could cost.

Get a custom quote

Pricing

Free Quota

Usage is free for the first six months, up to the following limits per Google Cloud Console project. This free quota applies even when billing is enabled, until the six-month expiration time.

  • Usage limits:
    • Predictions: 100 predictions/day
    • Hosted model predictions: Hosted models have a usage limit of 100 predictions/day/user across all models.
    • Training: 5MB trained/day
    • Streaming updates: 100 streaming updates/day
  • Lifetime cap: 20,000 predictions.
  • Expiration: Free quota expires six months after activating Google Prediction for your project in the Google Cloud Console.

Note that some Prediction API usage scenarios depend on Google Cloud Storage, which has different pricing terms from the Prediction API.

Paid Usage

Paid usage comes with a 99.9% availability service level agreement. Read more about the SLA here.

Monitor your usage on the Google Cloud Console. Any usage beyond the free usage quota will fail if you have not enabled billing for the project. Once you exceed the free quota, the following prices and limits apply:

  • Base fee: $10 monthly fee per Google Cloud Console project.
  • Usage fees:
    • Prediction:
      • 10,000 predictions/month: $0.00
      • 10,001+ predictions/month: $0.50/1,000 predictions beyond the initial 10,000
    • Training:
      • $0.002/MB bulk trained (maximum size of each dataset: 2.5GB)
      • 0-10,000 streaming updates: $0.00
      • 10,001+ streaming updates: $0.05/1,000 updates beyond the initial 10,000.
  • Usage limits:
    • If you intend to make more than 40,000 predictions/day, please contact us.
    • Google Prediction has a default limit of 2,000,000 predictions/day per project for your own models.
    • Hosted models have a usage limit of 100 predictions/day/user across all demonstration models. Developers can set higher paid usage limits for models they create. Please see each model's documentation for more details.

Important: These charges do not include any Google Cloud Storage fees that might be required to hold training data. Please see the pricing sheet for Google Cloud Storage to determine any additional costs that you might incur.