Level 5 Armor (RF4) – It doesn’t exist, but it should.
By now the NIJ’s new 0101.07 specification is on its way to becoming well established. To briefly review, its different levels are as follows:
NIJ Ballistic Protection Level |
Test Threat |
Ammunition Identifier |
Reference Velocity |
NIJ RF1 |
7.62x51mm M80 Ball NATO FMJ Steel Jacket 147 +/- 0/3 grain |
U.S. military supply or rounds meeting NATO specifications |
2780 ft/s (847 m/s) |
|
7.62x39mm MSC Ball Ammunition Type 56 from Factory 31 |
Factory 31 Type 56 Ammunition evaluated and meeting requirements in |
2400 ft/s (732 m/s) |
|
5.56mm M193 56 +/- 0/2 grain |
U.S. military supply or rounds meeting NATO specifications |
3250 ft/s (990 m/s) |
NIJ RF2 |
7.62x51mm M80 Ball NATO FMJ Steel Jacket 147 +/- 0/3 grain |
U.S. military supply or rounds meeting NATO specifications |
2780 ft/s (847 m/s) |
|
7.62x39mm MSC Ball Ammunition Type 56 from Factory 31 |
Factory 31 Type 56 Ammunition evaluated and meeting requirements in |
2400 ft/s (732 m/s) |
|
5.56mm M193 56 +/- 0/2 grain |
U.S. military supply or rounds meeting NATO specifications |
3250 ft/s (990 m/s) |
|
5.56mm M855 61.8 +/- 1.5 grain |
U.S. military supply or rounds meeting NATO specifications |
3115 ft/s (950 m/s) |
NIJ RF3 |
30.06 M2 AP 165.7 +/- 0/7 grain |
U.S. military supply or rounds meeting NATO specifications |
2880 ft/s (878 m/s) |
The RF1 and RF2 levels make sense. For reference’s sake, 0101.06 “Level III” only specified that armor plates need to stop six shots of 7.62x51mm M80 Ball. An 0101.07 RF1-compliant plate model now needs to show that it’ll stop three shots of 7.62x51mm M80 Ball, three shots of 7.62x39mm MSC, and three shots of 5.56x45mm M193 – though each threat is tested on a clean plate. (And repeated several times under different conditions, so that ultimately you need to submit several dozen plates of any given model for initial NIJ compliance testing.)
In a way, RF1 is an easier test than Level III – three shots rather than six – yet in a way it’s also substantially harder, as the M80 Ball is an easy threat to stop with a plate made entirely of pressed UHMWPE composite, whereas the steel-core 7.62x39mm threat is more of a challenge.
Well, I suppose the above point is debatable, but what I think cannot be disputed is the fact that RF1 is simply more comprehensive than Level III. In the latest FBI LEOKA table, 7.62x39mm is the second most common rifle threat:

And, in fact, more officers were murdered over the period 2013-2022 with 7.62x39mm (15) than were murdered with .308 (3), .30-06 (1), “7.62mm” (4), 7mm (1), 6.5mm (2), .270 (1), and .30-30 (1) combined. (13 in total.)
So, clearly, it’s sensible to have the most basic rifle protection specification encompass 7.62x39mm in addition to .308/7.62x51mm. In this case, from even a cursory look at the data, the more comprehensive specification is the better specification.
Another look at the table reveals that the most common threat by a wide margin is .223/5.56x45mm. The RF1 specification addresses this, as well, in mandating that plates stop lead-core M193 ball at (approximately) muzzle velocity from a 20” barrel. But RF1 addresses 5.56mm only partially – not completely. Steel-core 5.56x45mm M855 is very common, can safely be fired by many if not most rifles chambered in .223, and generally (though not always) penetrates light armor better than M193. Under test conditions, it’ll penetrate all plates and helmets made entirely of UHMWPE with no known exceptions.
The NIJ 0101.07 specification covers this in its RF2 level, which is simply RF1 + testing against M855.
Now, looking at the table a third time, I’m not seeing any threat that an RF2-rated plate won’t stop. I don’t believe that AP rounds or military EPR rounds have ever been utilized in a police shooting, nor have I heard of their use in the commission of any crime. So perhaps they’ll penetrate an RF2 plate, but they are uncommon.
This brings us to RF3. This is exactly the same as 0101.06’s old Level IV. It’s supposed to stop everything – supposed to cover “armor piercing” rifle threats in general. It does not do a complete job of this.
The Problem With Level IV / RF3
The basic problem is that the RF3 reference threat – the .30-06 M2 AP – is nearly 100 years old and of very simple design: Gilding metal jacket, lead-filled nose, tool steel penetrator.

(Image from Burkins, Matthew et al. (2017). A Potential Methodology for Evaluating Ceramic Quality. Proceedings of the 30th International Symposium on Ballistics. 10.12783/ballistics2017/17036.)
A plate that’ll stop .30-06 M2 AP will stop M855A1 and M80A1 with ease – but just about every modern AP round in 7.62mm, and sometimes in smaller calibers, out-performs all of those. 7.62x51mm M993 out-performs .30-06 M2 AP by a substantial margin. M993 is generally out-performed by 7.62x51mm Swiss P AP. And even the Swiss P AP, now decades old itself, is soundly beaten by today’s best 7.62mm AP projectiles. A plate rated to Level IV/RF3 isn’t reliably going to stop any truly modern AP threat.
The German VPAM ballistic standard makes some provision for this. VPAM Level 10 has plates tested against 7.62x54mmR B32 API – a threat with similar performance characteristics to the .30-06 M2 AP. (And in fact the .30-06 M2 AP was first envisioned as a B32 API surrogate!) The next level up, VPAM-11, calls for testing against M993. The VPAM-12 test threat is Swiss P AP.

This is sensible. It could be updated, but it’s basically comprehensive. The NIJ specification’s treatment of rifle AP threats, in contrast, cannot be called comprehensive or even adequate.
The Problem Compounded: Foreign Military Reliance on the Civilian NIJ Specification
Here’s where the problem gets ugly: The US Military has its own specifications for armor plates and doesn’t appear to utilize any of the NIJ’s specifications, but a lot of foreign militaries – including many NATO militaries – procure plates built to NIJ specs. They don’t use VPAM and don’t even necessarily have access to the details surrounding the US Military’s armor plate specifications. In a military context, where AP rounds might fly that have far more penetrative power than the old .30-06 M2 AP, Level IV/RF3 protection comes up short, but many military procurement arms are not aware of this.
The NIJ’s current system was built from the ground up for US domestic law enforcement – it’s not intended for military use – and for US domestic end-users it’s a good fit. But, frankly, it is also used by militaries around the world, where it’s less than adequate. It would therefore be good if the NIJ left some room for intermediate or advanced threats. The ideal “baseline” rating for a generic military-issue hard armor plate would be something like RF2+M855A1, and the ideal “AP” rating would be something broadly analogous to VPAM-12.
With that, we can re-jig the entire rating system.
RF1 (Level III): Rifle ball threats – 7.62x51mm M80, 7.62x39mm MSC, 5.56x45mm M193
RF2 (Level “III+”): Rifle penetrator threats – 5.56x45mm M855 (also 5.45x39mm 5N7)
RF2.5: Rifle EPR threats – 5.56x45mm M855A1, 7.62x51mm M80A1 (also 5.45x39mm 7N6/7N6M)
RF3 (Level IV): Steel-core rifle AP – .30-06 M2 AP, 7.62x54mmR B32 API
RF4 (Level V): Modern rifle AP – 7.62x51mm M993, Swiss P AP, similar threats.
As with VPAM, this would allow one rating system to encompass the entire range of small-arms ballistic threats.
So, as we see it, the NIJ really ought to consider an RF4/“Level 5” spec for armor. This should naturally be something that stops heavier small arms AP threats. Perhaps something built to VPAM-12. Perhaps something like the Colossus. That’s what Level 5 ought to be, and we already have it.