Tomorrow.io

Tomorrow.io (formerly ClimaCell) is a weather-intelligence platform whose Weather API has become one of the most feature-rich offerings in the category. Rather than simply reselling NOAA/ECMWF model output, the company runs its own proprietary forecasting stack that fuses public model data, satellite imagery, and non-traditional signals (including connected vehicles and wireless-signal attenuation) with AI/ML to produce hyperlocal forecasts. The API exposes 60+ data layers, far beyond core temperature/wind/precip into air quality, pollen, lightning, solar radiation, soil, maritime, aviation, fire and flood indices, across realtime, intraday, up-to-14-day forecast, and ~20 years of historical data, queried through a unified "Timeline" endpoint plus dedicated Realtime, Forecast, Historical, Maps (tile), and Route/monitoring endpoints. It is aimed squarely at developers and operationally weather-sensitive enterprises (aviation, logistics, energy, sports, government) who need richer parameters and per-point/polygon/polyline geometry than commodity weather APIs provide.

Commercially, Tomorrow.io splits into an API plan (priced on call volume) and a Platform plan (priced on seats/monitored locations/features). The free developer tier is genuinely usable for prototyping, 500 calls/day, 3 req/s, up to 5-day forecast, 1 alert and 1 monitored location, but the most differentiated parameters (pollen, lightning, air quality) and any meaningful scale sit behind "contact sales," which means the public pricing surface is thin and real costs require a sales conversation. This is the main friction point versus self-serve competitors like OpenWeather, WeatherAPI.com, or Visual Crossing, where paid tiers and per-call prices are published outright.

On reputation, Tomorrow.io is widely cited (including in its own and third-party "best weather API" roundups) for breadth of data and a polished developer interface, and it lists marquee customers (Delta, JetBlue, Uber, Ford, NFL, U.S. Air Force, National Grid). The clearest negative signal comes from its consumer weather app, where users on the App Store report notable temperature inaccuracies in some locations and slow support response, a reminder that "hyperlocal AI forecasting" quality is location-dependent. For API buyers the practical trade-off is rich capability and a 99.9% advertised uptime against opaque enterprise pricing and the need to validate forecast accuracy for their specific geographies.