If you’re looking to build a compact PC, whether it’s for gaming, work, or to use as a home theater (or a server), low-profile GPUs a must-have.
These more compact variants of their full-sized brethren though usually come with a performance sacrifice.
The GPU market is crazy at the moment, with heavy price-inflations being the norm and stock being hard to come by.
But that shouldn’t stop you from building a PC, even if it means you’ll have to work extra hard to find a new GPU that you can afford.
With that in mind, let’s dive into the details of the best low-profile GPUs out there.
What are low-profile GPUs?
Low-profile GPUs are more compact variants of full-sized models, made to fit into a smaller case.
Small form factor builds have become popular in the past couple of years, and low-profile GPUs are your best bet if you’re building one. There are a few things to keep an eye on in greater detail when shopping for a low-profile GPU – noise levels, power draw, and thermals.
With a low-profile GPU, you usually get one of two scenarios.
- You get a GPU that performs identically to the full-size model but will heat up more and make more noise.
- You get a GPU that performs slightly worse but may or may not be quieter and will be less demanding when it comes to keeping it cool.
Let’s take a look at both options.
Identical performance, worse thermals, and noise levels
The first option is usually a dual-fan version that is more compact than the conventional GPU.
The performance is identical, but those two fans will need to work harder to keep it cool because there is less surface for heat dissipation.
Manufacturers often deal with this by using a different fan design that moves more air but makes more noise.
These GPUs may use slightly less power than a conventional model, but the difference is usually negligible and you’ll still need a good PSU to keep them satisfied. You do, however, get the performance of a full-size model.
Slightly lower performance, good thermals, and low noise
The other solution is a single fan GPU, which is known as a true low profile model.
These single-fan GPUs are very compact, which means that there is less surface area for heat dissipation, and in addition, they only have a single fan, making cooling even more difficult.
They’re also almost always clocked lower than their full-size counterparts.
And with single-fan GPUs, one of two things can happen:
- You can have a high RPM (Revolutions-per-minute) fan that does a better job at cooling but makes a lot of noise.
- You could have a fan that spins at normal speeds and noise levels, but that also struggles with temperatures.
In the first scenario, the higher RPM, and the lower clock speeds compensate for the lack of surface area, and you still get decent performance.
In the second scenario, the fan is quiet, but since it can’t do too much in terms of cooling, you get higher temperatures and the clock speeds are limited and you’ll see occasional performance throttling.
Regardless of which kind of low-profile GPU you decide on, you will have to compromise.
This is to be expected, though, when building a compact system. But let’s take a look at what those compromises mean in day-to-day use.
Downsides of Low-Profile GPUs
If you’re buying a GPU for professional workloads that can take advantage of GPU acceleration, such as video editing or 3D rendering, the most notable downside you’ll need to deal with is the performance hit.
Of course, how much of a performance hit you take will depend greatly on what type of low-profile GPU you opt for.
If, for example, you don’t mind the extra noise, and do have room for a dual-fan GPU, the performance hit is going to be negligible in most scenarios. On the other hand, if you need the absolute smallest GPU, you’ll be dealing with low clock speeds and temperatures that might be uncomfortable in some cases.
GPU accelerated workloads depend on the GPU memory capacity as well, which is identical between a low profile model, and a full-sized one. But the clock speeds are just as important.
To add to this, while gaming on a low-profile GPU is certainly possible, you’ll struggle to get the performance of a full-size GPU.
These things aren’t all due to the lowered clock speeds – the design itself has a lot of impact.
You’ve probably heard of thermal throttling, which occurs when your GPU is reaching temperatures that are higher than the manufacturer would like them to be and underclocks the GPU to cool it down.
And with a compact design, this is more of an issue than it is with a full-sized one.
Why is this a problem? Well, because workloads like rendering will keep your GPU busy for extended periods of time.
Temperatures will keep rising, and you’ll likely see thermal throttling after a while. With workloads that only push your GPU for short periods of time, though, it’s not that much of an issue.
Who needs a low-profile GPU?
A low-profile GPU means compromise., We recommend that you only buy a low-profile GPU when the situation calls for it. What are some of those situations?
The first situation – building a small form factor PC. We’re talking about systems that use mini ITX or micro ATX motherboards, are packed inside a small form factor case, and don’t have that much room for components.
These builds are seldom recommended for professional purposes, but some of you who just want a good day-to-day system for general use will love them.
The reason is simple – they don’t take up a lot of space and can be hidden away easily.
The second situation is building a home theater PC. These are systems that usually go in a cabinet, or sit next to your TV, and don’t do much except for storing content and playing it on demand. What’s nice about these workloads is that usually, they aren’t too demanding in terms of hardware.
Your GPU will only need to render the content you’re playing, at the resolution you need. Of course, if you’re watching high-bitrate, 4K content, you want at least some performance from that GPU.
If you’re only building it for full HD content consumption, any budget low profile GPU will do the job.
Last but not least, you can compromise some portability for good performance.
This is why, rather than opt for a laptop, you can build a small PC that you can carry with you. In these situations, what you use the PC for will dictate what the best low-profile GPU for you is.
Choosing the right GPU
When it comes to choosing the best low-profile GPU for your needs, there are a couple of factors to consider. The two most important ones are:
- Size constraints you’re dealing with and
- Your workloads.
The use case is arguably more important, especially if you’re using your system to get some work done.
In this case, a more powerful GPU help you get more work done, so the investment will be worth it.
That being said, some of the more powerful GPUs just aren’t available in a small form factor, such as a single fan design. If you’re limited in terms of size, you’ll need to see what the most powerful GPU is that you can fit inside your case, and work from there.
Then there’s the power draw situation.
Small form factor power supplies aren’t readily available at high wattages, so you might need to get a GPU that draws less power to be able to run it without issues.
All factors considered, here’s what we recommend:
- Check your workloads to see whether you can benefit from a powerful model.
- See how much case space you’re working with, and what the biggest GPU is that you can fit in it.
- Consider your power supply constraints, and see what kind of GPU you can run with the PSUs available to you.
Once you’ve done all of this, you can see what the most powerful GPU is that satisfies all of the above and fits your budget, and pull the trigger on that purchase.
Low profile GPUs, as a product, are a compromise. They aren’t meant to be the most powerful option out there, and they’re oftentimes budget-oriented.
This is why you won’t find a single-fan RTX 3000 series or an RX 6000 series GPU. This is also why there are still many low-profile GPUs that use GDDR3 memory – more powerful models moved on from that type of VRAM a long time ago.
Low-profile GPUs aren’t a bestseller for any manufacturer, which is why they’re put on the back burner when it comes to new GPU releases.
In the list below, we include some modern options, but there are some older ones, too. Let’s take a look at them.
The best low profile GPUs you can buy today
The GTX 1080 is, at this point, an almost five-year-old GPU. Why is it still relevant? Because it’s had its price reduced a couple of times, and Gigabyte’s Mini variant is one of the smallest GPUs out there that can get you this kind of power.
There are 8GB of GDDR5X memory, as well as an OC mode that can push the boost clock up to 1771Mhz, which is pretty decent.
What’s great is that accessing that OC mode is a single click away, so you don’t have to go through the trouble of manual overclocking.
On the outside, the 1080 Mini’s dimensions come in at only 17cm in terms of length, which is impressive.
A large part of that is thanks to the 90mm fan, which is smaller than most competitive models out there. If you don’t mind the fact that this isn’t exactly new, it’s certainly a viable option when you’re limited in space.
PowerColor’s ITX variant of the RX 5600 XT is truly a great choice. Yes, it might produce more heat than you’d like it to, but that isn’t much of an issue unless you keep it under load for extended periods.
Spec-wise, you’re looking at 6GB of GDDR6 memory, as well as 2304 stream processors.
The boost clock isn’t particularly impressive at a 1620Mhz maximum, but it gets the job done unless you need top-notch performance.
You do get two DisplayPort connectors, and one HDMI, which is impressive for a GPU that’s only 17.5cm in length and compact overall.
You’ll need at least a 500W power supply to run it, though, and it requires you to plug in a single 8-pin connector for power. That sounds reasonable if you ask us.
Not everyone needs a huge amount of performance from their low-profile GPU. Sometimes size is the main constraint, which is where something like MSI’s LP OC variant of the GT 1030 comes in. This GPU is a single-slot, low-profile model with a very compact cooler, making it perfect for an HTPC.
In terms of performance, you’re looking at 2GB of GDDR5 memory and a clock speed of 1518 MHz. What’s nice is that this kind of configuration can easily be used for 4K video playback on an external monitor or TV.
There is a single HDMI and single DisplayPort at the rear, so you could potentially run two monitors if needed. The extremely compact fan is near silent, and there is no need for a an external power connector either.
Wrapping things up with another RTX 3000 series GPU, this time it’s the RTX 3060 Eagle OC by Gigabyte.
Similar to the RTX 3070 we mentioned earlier, you get 2nd generation ray-tracing cores and 3rd generation Tensor cores, for optimal use of technologies such as ray tracing and DLSS for competitive gaming and fast 3D Rendering.
What’s impressive is that you get 12GB of GDDR6 memory, which is great for GPU-accelerated software that requires high amounts of memory.
The boost clock is a respectable 1807Mhz, and you’re looking at a total of 3584 CUDA cores.
The GPU does come with four display connectors in total, which is to be expected with a dual-fan design, as well as a PSU recommendation of 550 watts, and a requirement for a single 8-pin power connector.
If you’ve got a bit of room in your case, and want the latest generation performance, ZOTAC’s GeForce RTX 3070 Twin Edge OC is a great choice. With current market prices, it is expensive though. But it’s the best choice for great performance in a compact package.
The Twin Edge OC comes with 8GB of GDDR6 memory, as well as Nvidia’s Ampere architecture that brings along 2nd generation Ray Tracing cores and 3rd generation Tensor cores.
This all results in an excellent performance, not just when gaming, but also when editing and rendering videos, and any kind of 3D design.
This GPU’s boost clock is a pretty decent 1755Mhz, which when paired with the 5888 CUDA cores and the 14 Gbps memory clock, ensures fast render times.
You will need a 650W power supply to run it, and the dual IceStorm 2.0 fans do make a bit of noise. Note that this is a dual-slot GPU, so make sure you have enough room for it.
While single-fan designs are usually more compact than this, EVGA has done a good job of keeping the RTX 2060 compact, yet powerful. It has plenty of surface area for heat dissipation, and the fan is surprisingly quiet.
In terms of performance, you’re looking at a pretty standard RTX 2060 configuration – 1920 CUDA cores, 6GB of GDDR6 memory, and a maximum boost clock of 1680Mhz. This is quite sufficient for most use cases, and that does include GPU-accelerated video editing, too.
What’s neat is that you have a maximum support for three monitors, which comes in handy if you’re running a multi-monitor productivity setup. To add to this, you can run it on a 500W power supply.
It does have a triple-slot thickness, so watch out if you have enough space available for it.
Conclusion – is a low-profile GPU worth it?
The compact GPUs we recommend in this article perform reasonably well in SFF (Small-Form-Factor) cases that are well-ventilated.
In the end, though, low-profile GPUs are a niche market for PC-Builders that, first and foremost, have to work with size constraints and can live with performance hits due to thermal throttling or lower clock speeds.