You could do a simple PIVOT

Note: The WHERE is optional, but may be more performant

Example

Select * 
 From  (Select Md5
         From  YourTable
         Where md5 in ('xxxxx1', 'xxxxx2')  -- << Optional
       ) src
 Pivot ( count(Md5) for Md5 in ([xxxxx1], [xxxxx2]) ) Pvt

Results

xxxxx1  xxxxx2
1       1

The easiest way to do this is to start with a table of the MD5 hashes to check, and left-join the file table.

You can use a VALUES constructor

SELECT
  v.Md5,
  f.Id   -- this column is NULL if not matched
FROM (VALUES
    (0xxxx1),
    (0xxxx2),
    (0xxxx8)
) v(md5)
LEFT JOIN file f ON f.md5 = v.md5;

Or, depending on how you pass the data in, you can use a table variable, temp table or Table-Valued Parameter.

DECLARE @values TABLE (md5 binary(32) PRIMARY KEY);
INSERT @values (md5)
VALUES
    (0xxxx1),
    (0xxxx2),
    (0xxxx8);

SELECT
  v.Md5,
  f.Id   -- this column is NULL if not matched
FROM @values v
LEFT JOIN file f ON f.md5 = v.md5;

The results would be something like this