Cloudy,much the same tomorrow
NWS, the American and the European model put San Francisco's high today between 68° and 73°, a 5° split. Tomorrow looks partly cloudy around 68°, with a 1% chance of rain.
San Francisco, CA. Below we compare what NWS, American, European forecast for San Francisco right now and flag exactly where they disagree, then break down San Francisco's weather month by month and which forecast model has been most accurate here.
We grade every model against the weather that actually happened, so you can see which forecast model is most accurate in United States before you trust any single number.
Here is San Francisco's typical weather in each month: the average daytime high and overnight low in °F, how each month feels, and the wettest month marked. Use it to plan any trip to San Francisco by season.
San Francisco runs from warm summers, with October highs near 73°F, to cool winters, when February lows sit around 45°F (a swing of about 28°F across the year). January is usually the wettest month, and San Francisco sees roughly 33 inches of precipitation annually.
The most comfortable time to visit San Francisco is usually May, June, July and August, when daytime highs sit in the pleasant 60-78°F range and the heaviest January rain is behind you.
The CheckWeather Score blends temperature, wind and rain accuracy — each measured against ERA5 (the recorded weather) over the last 30 days across 1999 cities. Lower error means a higher score. See which forecast model is most accurate in United States, how we measure forecast accuracy, or which model each weather app uses.
Bottom line: over the last 30 days the European model has been the most accurate forecast for San Francisco, scoring 90/100 on our CheckWeather Score, ahead of the American at 72. See the full US accuracy scoreboard for exactly how every model is tested.
The models often agree on temperature and still split on rain — and rain is usually what decides whether a plan survives. A wide rain range means the shower is genuinely uncertain, not that the forecast is wrong.