GribStream

About GribStream

GribStream is a weather data API providing access to live forecasts, historical forecast archives, and reanalysis actuals through a single interface. The API accepts queries for coordinates, time ranges, calculated fields, and condition filters, returning data as CSV, JSON, or NDJSON. It covers global and regional forecast models and datasets including NBM, GFS, HRRR, RRFS, GEFS, IFS, ERA5, and AIGFS, along with satellite (GOES/GMGSI), radar (MRMS), and surface analysis (RTMA/URMA) products, totaling 89 datasets in its full catalog spanning forecasts, ensembles, seasonal products, hydrology, and atmospheric composition.

The API supports features such as automatic best-run selection, a model-run-time cutoff parameter (asOf) for backtesting against historical forecast cycles, derived-field computation, condition-based search, and access to full ensemble members and statistics. It is designed to handle requests for tens of thousands of points in a single HTTP call.

The product is aimed at developers building applications, machine learning pipelines, dashboards, and research workflows that require weather data, including customers in sectors such as professional sports analytics, ski-resort forecasting, and aviation weather. GribStream offers open-source clients, documentation, and a free API token to get started, along with expert support from engineers.

Key features

• Query global and regional weather forecasts, ensembles, satellite/radar data, and reanalysis actuals via one API
• Returns data as CSV, JSON, or NDJSON for coordinates, time ranges, and calculated fields
• Backtest by replaying older forecast cycles against the same coordinates and variables
• Derived-field expressions and condition-based filtering instead of raw grids
• Full ensemble member and statistics support
• Fetches tens of thousands of points in one HTTP request in seconds

Frequently asked questions