Dimension scores are derived from public data and fields; weighted into the composite. Reference only.
GribStream is a historical weather forecast API—or more precisely, a cloud database for weather model data. It unifies data from NOAA, ECMWF, Taiwan CWA, and AI forecasting systems into a queryable SaaS interface. Developers can retrieve historical or real-time forecast data by coordinates, grid, time range, model run time, and variables, without having to download, archive, and parse large volumes of GRIB data themselves.
Its main value lies in “database-style” weather queries. The Timeseries endpoint returns, for each valid time, the shortest matching forecast lead time, making it suitable for building nowcast or backtesting series. The Runs endpoint is used to fetch data by model run batch. The API supports CSV, JSON, and NDJSON, as well as gzip compression, explicit time lists, asOf backtracking cutoffs, min/max lead time, coordinates, and regular grids. Advanced features include server-side expression calculations, derived fields, conditional filtering, and ensemble member selection, allowing users to filter for specific weather conditions instead of only returning raw grid values.
Model coverage is strong, including HRRR, GFS/GDAS, GEFS, CFS, RTMA/URMA, NBM, RAP, NAM, ECMWF IFS/AIFS, GraphCast, FourCastNet, AIGFS/AIGEFS, and Taiwan CWA WRF. The API uses Bearer Token authentication, and the documentation provides request headers, request body structure, variable objects, expressions, filters, response fields, error codes, and cURL examples. It also mentions an open-source Python client on GitHub. Overall, the documentation is developer-friendly, though variable selection requires exact identifiers from the model catalog, so teams without a meteorological background will still face a learning curve.
Pricing is transparent and includes a free plan. Pro ranges from $9.90/month for 48,000 credits/day to $244.70/month for 1,536,000 credits/day. Credits are calculated based on the number of returned valid times, parameters, and coordinate batches; cache hits are charged at only 10%, which is favorable for bulk but reusable queries. The strengths are broad model coverage, powerful query capabilities, flexible output formats, and controllable costs. The downsides are that the main content does not specify SLA, data latency commitments, payment methods, company location, or self-hosting options, and the only officially specified SDK is Python.
GribStream is suitable for teams in energy, agriculture, insurance, aviation, logistics, research, and AI applications that need historical forecast backtesting or multi-model weather features. Access from China cannot be determined from the available content alone. Before production use, it is recommended to test network connectivity, latency, registration, and payment flow. If access is limited, alternatives to evaluate include OpenWeather, Meteomatics, Tomorrow.io, Visual Crossing, or building an in-house pipeline directly on NOAA/ECMWF Open Data.
⚠ This review is compiled from public sources and does not constitute a purchase recommendation. Verify all facts on the vendor's official site. Verify on gribstream.com official site.
gribstream.com is an United States API & Data provider. TG4G tracks its product information, with monthly pricing from $9.90, an overall rating of 7.0/10, and a China-accessibility score of Workable. Click "Visit Official Site" to reach gribstream.com directly.