If it allows CR123s then it almost certainly does not have a low voltage cutoff for 16340s, since that cutoff would prevent the usage of CR123s. To get around that you have to do something fancy like HDS does where it detects the voltage and decides whether the battery is 16340 or CR123. But if they implemented that feature they'd probably advertise it.
Not really very "fancy" in '21 - more like 'old hat' to me now, but perhaps I'm just spoiled. I
expect this feature in a modern, quality design. Both of the two Fenix 1xCR123A/16340 lights I use (PD25 & RC09) employ this technique for enabling their 'Low Voltage Warning / Indication' feature, and as I recall they're about 5 years old now, and cost
far less than the newly updated light under discussion here. It appears to work very well in practice / my experience, I would also add. That's why I was surprised (actually shocked) to find that this new light (apparently) lacks this feature. One need not spend big money to get good features these days, but that said, when one does spend 'good money', they should expect good features.
Coincidentally, I was pleasantly surprised when, having slipped a bit and lost track of my SOC just last week (yes, it can happen), I discovered that my ET D25C MKII also incorporates 'low voltage warning indication', which was likely added with the MKII update for, among other improvements, 'proper' 16340 (lithium ion) support (sound familiar?). As far as I can determine, that 'feature' is not documented anywhere in the product documentation / specifications info, but it's there, and that was a very pleasant surprise for me! Edit: 'Pleasant' meaning it didn't punish me for my error by leaving me in sudden and sustained darkness, which in my use environment(s), can seriously compromise my personal safety.