im sure it has its own theory and ideas to gain an advantage, but white if white is careful then they will keep the advantage and later on use black's exposed king to their advantage.
Thoughts on the Duras Gambit

Latvian Gambit is better, in my opinion!
It's a much worse version of the King's and Latvian Gambit and if I know how to count White wins a tasty pawn, 2. exf5 1-0
Okay how do I sum this up... There is a fundamental problem with this. After exf5 white is not only a pawn up, but has a nasty thorn in black's position, causing the following problem:
The e-pawn can't move, because of en-passant. So the great center you are dreaming of can only consist of d5 + c5. Now you only want to expand on the queenside if you plan on castling kingside. The problem: How do you wanna castle kingside? To achieve that you would have to get your bishop out first, which requires yet again a move that the f5-thorn-pawn can take and white would be a healthy pawn up.
Now the only way to solve this problem is to get rid of the thorn, which is not only very hard to achieve, but moreover very time consuming and dangerous as black needs to constantly parry the threats of white's iniative along the open diagonals. In general you shouldn't move the f-pawn before being castled.
To e4 there are four very good responses: The French, the Caro-Kann, the Italian and the Sicilian. Pick one of those.
@Marcyful By "Italian" I meant all the e5-Games together, I just wanted to give it a fancier name. You can of course also play the Scotch, Spanish or Russian, they are all solid openings.
Isn’t the düras gambit the king’s gambit but for black
(The purpose of this is the same with the King’s Gambit, once black kingside castles the rook will be immediately active on the open f-file, and black can make the king safer by playing Kh8)

Think LM and Pizza did a pretty good job explaining the shortcomings.
I think I'd put it this way: if you want to gambit your f-pawn you have to play your e-pawn first.



the Duras gambit gambits the f in order to lure the e
out of the center. while black gains a good center this comes at the cost of king safety, as the f and g
s are important defenders of the black king

Okay how do I sum this up... There is a fundamental problem with this. After exf5 white is not only a pawn up, but has a nasty thorn in black's position, causing the following problem:
The e-pawn can't move, because of en-passant. So the great center you are dreaming of can only consist of d5 + c5. Now you only want to expand on the queenside if you plan on castling kingside. The problem: How do you wanna castle kingside? To achieve that you would have to get your bishop out first, which requires yet again a move that the f5-thorn-pawn can take and white would be a healthy pawn up.
Now the only way to solve this problem is to get rid of the thorn, which is not only very hard to achieve, but moreover very time consuming and dangerous as black needs to constantly parry the threats of white's iniative along the open diagonals. In general you shouldn't move the f-pawn before being castled.
To e4 there are four very good responses: The French, the Caro-Kann, the Italian and the Sicilian. Pick one of those.
ewww, nah, B00 is wheres at. Owens and nimzowitsch boi!
One day I was thinking "what if black gambited their f-pawn in exchange for better center control?" And so I stumbled upon the duras gambit. I was wondering what people's thoughts were on it. (For those who don't know what I'm talking about, here)