I say number 3 is standard, but I chose number 7 (all of the above, except for 6).
1: with = wif
Speakers of this particular pronunciation (e.g. Ebonics?) replace the voiceless inter-dental (tongue between the teeth) /th/ with a voiceless labio-dental (lip between the teeth) fricative /f/.
2: with = "th" like in thin
as well as "th" as in the
Speakers of this particular pronunciation include North Americans and Europeans, including the me. Voiced /th/ and voiceless /th/ are in free variation. Speakers use both, even I use both.
3: with = th = s (but this i think is definitely wrong)
Speakers who pronounce /th/ as [s] do so because this voiceless alveolar fricative ([s]) is the closest sound to /th/ in their native sound system.
5: with = th = d
Speakers of the pronounciation include the populace of Da Bronx! as has been already noted. The voiced inter-dental (tongue between teeth) /th/ is replaced by the voiced dental (tongue inside oral cavity) [d].
6: = none of these
7: = all of the above, except 6.
8)
with 
Linear Mode
