As Webster 1913 says, the Farad is the measure of Electrical capacitance. To put the technical definition slightly more clearly, if you apply a charge of 1 coulomb (ie 1 ampere for 1 second) to a fully discharged capacitor, and the voltage across the capacitor on completion is 1 volt, the capacitor has a capacitance of 1 Farad.
What it doesn't say, though, is that a Farad is big. Really big. A one farad non-electrolytic capacitor would have a surface area in the orders of square kilometers. Even a one farad electrolytic capacitor has a surface area of about 13 square meters, although this can be squashed into a small tin can sized unit.
Typical values for non-electrolytic capacitors are measured in nanofarads or even picofarads, with some up to large fractions of microfarads. And these are all only for comparatively low voltages - up to perhaps 50v or 100v. On the other hand, these are also very cheap, with prices in the orders of a few pence each.
Then we have electrolytic capacitors. These have much higher ratings, but are polarised (have permanent positive and negative electrodes), and also tend to leak charge faster. However, they are usually expressed in microfarads. They are also more expensive - a 10,000 microfarad (I've never seen the term "milifarad" used), with a 63v rating will set you back close to £10.
There are also speciality capacitors with even higher ratings, up to and including 1 farad. However, these are usually only capable of about 5v, and are used instead of batteries for CMOS memory retention. These are also electrolytic.