I am introducing Xbench to my team and downloading a trial version.
We use XTM for our CAT tool. The glossaries that we use are imported into XTM.
I didn’t see XTM as an input format, but the caption did read “Zillion input formats” so perhaps XTM just wasn’t listed.
Can you recommend which version of Xbench would best fit my needs? Also, is there a demo/training that we can take in order to know how to use this tool?
If I’m not wrong, XTM is able to export XLIFF files that can be loaded by Xbench (or at least that was the case with XTM 7.5 when we checked it a couple or three years ago).
Xbench can also read XLIFF:doc files, which I believe is exported by XTM, but I think that was only for projects specifically configured to export XLIFF:doc files (perhaps this has changed).
For example, to process an XLIFF file with Xbench 3.0, you can right-click on the file in Windows Explorer and choose Run QA in Xbench.
To quickly learn Xbench, we strongly recommend checking out our YouTube Channel.
It contains several focused 5-minute demo videos that introduce features and provide useful hints (even for experienced Xbench users).
Perhaps to start, I would recommend following the Getting Started with Xbench playlist of 5-minute videos, which in barely 30 minutes give a very good overview of main Xbench features.
Integration with CAT tools is a top priority item for Xbench. Any opportunities to integrate with any tool are expedited.
We already have quite neat integrations with two online CAT tools similar in functionality to XTM: Transifex and Memsource (there are videos for both on our YouTube Channel).
If XTM were interested in facilitating a tight one-click integration such as the one we have for example with Studio, Memsource, memoQ or Transifex, we would give very high priority to the task of implementing such functionality. I suppose that XTM will be more prone to facilitating such integration if key customers keep asking them for Xbench integration.
As I checked, Xbench supports XTM’s XLF files, but sees all kinds of matches as 0% (no match), which makes the parameters “Only New Segments”, “Only 100% Segments”, “Exclude ICE Segments” useless.
Only after running XTM’s XLF through another tool (e.g. importing to memoQ and exporting as bilingual), Xbench sees the matches correctly, but such process is a waste of time 
This happens because our parser does not recognize the XTM XLIFF dialect yet. Probably XTM added a few more attributes to their XLIFF format since we last checked it a couple years ago.
It would be great if you could submit a bug with a sample XTM XLIFF file to our Contact Support page, as we would be able to look at the new attributes so that they are recognized out of the box in next Xbench build.
Thank you. I submitted a bug report.