I wonder the same thing myself. It's suboptimal because you forfeit the center but you don't give up any material that you weren't going to lose anyway.
Why is the Noa Gambit a gambit if it does not lose material?
I played against it a lot at my level and lichess' stockfish engine thinks every variation in the noa gambit is winning with black with perfect play, because the knights are very badly placed.

In the Noa Gambit, you aren't sacrificing pawns, but you are sacrificing the bishop pair in an open position,center control, and the initiative. That's worth more than one pawn.

You sacrifice your position it’s terrible. The bishop sacrificed to open the king would like to be a member of the attack. The sacrifice is usless otherwise. Giving the knight is a lot better

Of Course, your playing a Gambit!
You just gave up a Center Pawn + Bishop for your opponents Flank Pawn + Knight.
Black has an advantage for sure.
White will have to pray for draw!
Nc3 isn’t the correct move.
—————————
Best moves are Ng5 or d3.
Good moves are d4 or c3.
You sacrifice your position it’s terrible. The bishop sacrificed to open the king would like to be a member of the attack. The sacrifice is usless otherwise. Giving the knight is a lot better
good point, only knights sacrifice on f7.
Of Course, your playing a Gambit!
You just gave up a Center Pawn + Bishop for your opponents Flank Pawn + Knight.
Black has an advantage for sure.
White will have to pray for draw!
Nc3 isn’t the correct move.
—————————
Best moves are Ng5 or d3.
Good moves are d4 or c3.
Yeah I get the bad italian 4 knights position very very often from playing the petrov, when the opponent realize that the Bc5 Italian doesn't work because e4 is under attack, so they defend with Nc3 first before trying the Italian.
Bb5 in the 4 knights (the 4 knights spanish) is interesting as it is used by Carlsen when he is playing white to make a quick draw against Pragg when he was sick. He also drawed Ding Liring with it. https://lichess.org/35EuEaWc/black#50
I wonder who Noa is anyways?

It is terrible because 1) it allows Black total central control 2) surrenders the bishop pair early on for no good reason (Black's king isn't really exposed, nor his development is seriously hindered).
After 1.e4 e5 2.Nf3 Nc6 3.Bc4 Nf6 4.Nc3 Nxe4 5.Bxf7+? Kxf7 6.Nxe4 d5 white should seek to minimize the damage with 7.Ng3, when he is "just" much worse, instead of following the "patzer sees check, patzer gives check" principle: After 7.Neg5+? Kg8 he is lost already, as both his developed pieces are very poorly placed.
The following game is probably a "pre-arranged loss" but still very indicative about how bad white's position can become.
It is terrible because 1) it allows Black total central control 2) surrenders the bishop pair early on for no good reason (Black's king isn't really exposed, nor his development is seriously hindered).
After 1.e4 e5 2.Nf3 Nc6 3.Bc4 Nf6 4.Nc3 Nxe4 5.Bxf7+? Kxf7 6.Nxe4 d5 white should seek to minimize the damage with 7.Ng3, when he is "just" much worse, instead of following the "patzer sees check, patzer gives check" principle: After 7.Neg5+? Kg8 he is lost already, as both his developed pieces are very poorly placed.
The following game is probably a "pre-arranged loss" but still very indicative about how bad white's position can become.
Yeah Ne5+ just forces black to move the king to a safer spot, which he was going to do anyways.
This opening does not lose material.