No way built in that I know of. However you can delete the default Quantiles segment and replace it with a custom made one that shows the quantiles you are interested in. The following bare bones example suggests how this might be done.
There is probably a clever way to do this but here is the outline of a brute force method. The idea is simply to create a matrix version of the table which contains each value as many times as indicated by the frequency column and then to compute the desired quantiles from that matrix. Something like this code would be combined with the previous example in this thread.
style="font-size: 10pt; font-family: "Courier New"; color: green;">/* Assumes a table with first column as Y and second column as Freq
style="font-size: 10pt; font-family: "Courier New"; color: green;"> style=""> Expands Y column to a single column matrix with each original Y
style="font-size: 10pt; font-family: "Courier New"; color: green;"> style=""> value repeated Freq times */ style="font-size: 10pt; font-family: "Courier New"; color: black;">
Summary didn't appear to honor a column with the frequency role for quantiles when I tried it, though it does for means. Perhaps I should say that the results I got using summary on a column where another column with a frequency role existed differed from the results I got using summary on the expanded table. It's possible that the various quantile functions in JMP don't all use the same algorithm to deal with ties. As time permits I'll recheck using summary as that would be preferable.
Ok here's what I tried. I opened the Fitness data and ran Summary for Age which gave me a table of ages and their frequencies (N Rows which I renamed Freq). I gave Freq the frequency role. I ran Summary for Age in the original table and ran it again in the summary table. I chose several quantiles. I found that the 25th and 75th quantiles did not agree though the means did. The quantiles for the summary table were the same as if I had not specified a frequency column.