There is much information about WW2 German weapons out there- much of it is in excellent reference books such as Hitler's Garands and Backbone of the Wehrmacht- however, these are only guides, and not everything in them is correct. Case in point is 1945 date K98k production numbers. It is widely recognized that Backbone of the Wehrmacht is incorrect in it's total production figures for 1945 manufactured 98k's. Below is the best I have seen it put into context: (information used with author Eric Davidson's permission, my additions in orange). From Gunboards.com Mauser K98 Forum (member name Pisgah): (in regards to Backbone of the Wehrmacht production estimates for 1945 dated rifles) As they're estimates, they're all off to some extent. His numbers from 1945 are the farthest off of any of the years due to all of the changes which were instituted late war by the different manufacturers. 
Here are examples from 1945 including Law's estimates, and why they're wrong: 1.
World War II Nazi K98 Mauser 'ar/42' Code Bolt Action Rifle with Bayonet. Serial number obfuscated. What is a good website to use to help with identification of a model 98 German Mauser, such as the proof marks and serial number dating.
Mauser Oberndorf (byf/svw45)-Law estimated 205,591 based on the known high svw 45 at the time being numbered 5593b. Law ignored svwMB production as many consider this code to be all French postwar manufactured. I think this belief is unlikely, but will deal with just byf 45 and svw 45 for the sake of comparison. The old KCN ran a serial number study for 1945 Mauser Oberndorfs in 1997. The high number found in that study for the no suffix block of byf 45 was 59,467.
The high number a block svw 45 found was 16205a. The high number found for svw 45 in the b block was 5593b. So, actual production for byf 45 and svw 45 was less than 83,000. I say 'less than' because some of the byf 45's and svw 45's were rejected by the Waffenamt inspectors and never completed during the war (despite having been assigned serial numbers) and were eventually completed by the French postwar. Law didn't account for the rejected rifles and didn't account for the fact that Mauser Oberndorf never completed any of the 5 digit number blocks of rifles (leading many to believe that the serial number blocks at Mauser Oberndorf corresponded to months of the year from 1943 until the end of the war).
Gustloff (bcd45)-Law estimated 91,679 rifles based on the known high serial number of bcd 45 being 91679. The problem is that bcd 4's are known to exist with serial numbers all the way up to the 90 thousand range. When found with five digit serial numbers and in original condition, bcd 4's have late features--semi-Kriegsmodell, complete or partial phosphate finish, rougher machining marks, and fewer serial numbered parts.
Based on the fact that these features didn't show up at any manufacturer until late 1944, it's pretty reasonable to assume that Gustloff mixed bcd 4 and bcd 45 receivers in their 1945 production. The best I can offer you with this code is a guess. Electrochemistry by carl h hamann pdf file.
I see 4-5 (or more) bcd 4's with the five digit serial number for every bcd 45 I see. Part of this is certainly due to the fact that collectors know bcd 45 is uncommon and don't sell them often. So, I would guess that the actual number of bcd 45's was around 20-25,000.