GridStatus.io Hosted API

• Python client for accessing the GridStatus.io Hosted API.
• Browse all available datasets in our Data Catalog.

Installation

gridstatusio supports Python 3.10+. Install with uv or pip.

uv pip install gridstatusio

Getting Started

• Sign up for a Grid Status account and get your API key from the Settings page
• Set your API key as an environment variable: export GRIDSTATUS_API_KEY=your_api_key or pass to the client with client = GridStatusClient(api_key="<your_api_key>")
• You're now ready to start querying. List datasets with:

from gridstatusio.gs_client import GridStatusClient
client = GridStatusClient()

data = client.get_dataset('ercot_fuel_mix', limit=100, start='2025-01-01', end='2025-01-02')

• To see all available datasets, use client.list_datasets() or check out the complete Grid Status catalog at https://www.gridstatus.io/datasets

• NOTE: the Grid Status API has a 1 million rows per month limit on the free plan. This limit is very easy to exceed when querying data, especially real time prices.

  • Make sure to add limit to all of your get_dataset calls to avoid quickly exceeding the limit.

• For more detailed examples, check out this notebook: Getting Started

• Other notebooks in the Examples directory:

  • Finding Hubs and Zones in Pricing Data
  • ERCOT Pricing Data
  • CAISO April Net Load Analysis
  • Stacked Net Load Visualization
  • Resample Data to Different Frequencies

Checking your API usage

usage = client.get_api_usage()

• This shows the limits for your API key, the start and end of the current usage period, and the API usage in the current period. Note a limit of -1 means no limit.

Retry Configuration

• The Grid Status API has rate limits that restrict the number of requests that are allowed each second, minute and hour. If rate limits are hit the client will automatically retry the request after a delay. You can configure the maximum number of retries using the max_retries parameter when initializing the client. If you find yourself hitting rate limits, you may need to add a delay between your requests. The Grid Status Pricing Page contains more details on specific rate limits.
• The client retries failed requests due to rate limits (429), server errors (5xx), and network issues using exponential backoff. You can customize retry behavior:

client = GridStatusClient(
    max_retries=3,        # Maximum retries (default: 5)
    base_delay=1.0,       # Base delay in seconds (default: 2.0)
    exponential_base=1.5, # Exponential backoff multiplier (default: 2.0)
)

The retry delay follows the formula delay = base_delay * (exponential_base ** retry_count).

Retries are useful when:

• You're making pagination-heavy requests and risk hitting short-term rate limits
• A request fails due to a temporary server error
• A network issue or timeout interrupts the request

To disable retries entirely, set max_retries=0.

Version Check

The client checks for updates to the library when it is imported. It does this by making a call to the library on PyPI. For certain applications, this call can be problematic, and can be disabled by setting the environment variable GSIO_SKIP_VERSION_CHECK to true.

export GSIO_SKIP_VERSION_CHECK=true

Open Source

If you prefer to use an open source library that fetches data directly from the source, you can check out this github repo.

Get Help

We'd love to answer any usage or data access questions! Please let us know by emailing us at contact@gridstatus.io