Here is a note about video cards for use with Stable Diffusion and some of their characteristics that need to be taken into account for good generation speed.

Long story short, I have a GeForce GTX 1660 Super, which is too slow for playing around with Stable Diffusion. Generating a basic 512×512 image with the SD1.5 model takes around 20–30 seconds – not too bad. However, when using inpaint mode, the time increases to several minutes. It’s fine for experimentation but requires a lot of waiting. So, I decided to review the market to understand what is currently available (as of February 2025) within my budget. Like most people, I don’t want to overspend, so I set my budget limit at 1,250 euros (roughly the same in USD).

The first step was to determine the key characteristics to consider when researching video cards.

First of all, I prefer Nvidia, so I didn’t look at AMD.

The key factors include:

• Generation (e.g., 3xxx vs. 4xxx)
• VRAM amount (8GB – 24GB)
• VRAM frequency
• CUDA core count
• GPU clock speed
• Other factors: card size, number of ports, number of fans, LED backlighting, power consumption, etc.

And, of course, the king – price.

My current video card

Here are the specs of my current GPU:

• Name: GeForce GTX 1660 Super
• Generation: 16xx
• VRAM: 6GB GDDR6
• VRAM frequency: 14GHz
• CUDA Cores: 1408
• GPU clock speed: 1.53GHz

What do we need for Stable Diffusion?

• VRAM – This is used to load models and process images. The VRAM consumption during image processing depends on resolution – higher resolutions require more VRAM.
  • SD1.5 (with a 2GB model file) easily maxes out my 6GB VRAM when generating 512×512 images.
  • SDXL needs even more since the model itself is around 6GB. The recommended minimum for SDXL is likely 12GB, given that it was trained on 1024×1024 images.
• CUDA Cores – More cores directly affect generation speed.
  • For example, if my current card has 1,408 CUDA cores and a new one has 4,500, I can expect roughly 3× faster generation – cutting the time from 20 seconds to about 7 seconds (or even faster).
• Generation – Each new GPU generation introduces better technology, higher memory bandwidth, and improved efficiency.
  • Choosing a newer generation is always preferable if all other specs are comparable.

GPUs that fit my requirements

I checked amazon.it and found the following cards that match my filter. I also included some top and popular cards like 3090, just for comparison:

Model, Nvidia RTX	VRAM DDR6x marked with x	Cuda cores	Cuda cores, x from 1660	VRAM freq	GPU freq	Avg Price, eur
GTX 1660 Super	6GB – 6	1408	1.00	14 Gbps	1.53 Ghz	250
3060Ti	8GB – 6	4864	3.45	14 Gbps	1.41 Ghz	450
4060Ti	8GB – 6	4352	3.09	18 Gbps	2.54 Ghz	500
3060	12GB – 6	3584	2.55	15 Gbps	1.32 Ghz	300
3080	12GB – 6x	8704	6.18	19 Gbps	1.44 Ghz	1000
4070	12GB – 6x	5888	4.18	21 Gbps	2.47 Ghz	650
5070	12GB – 7	6144	4.36	28 Gbps	2.51 Ghz	780
4070 Super	12GB – 6x	7168	5.09	21 Gbps	1.98 Ghz	850
4070Ti	12GB – 6x	7680	5.46	21 Gbps	2.47 Ghz	1200
4060Ti	16GB – 6x	4352	3.09	18 Gbps	2.54 Ghz	550
5070Ti	16GB – 7x	7680	5.45	32 Gbps	2.67 Ghz	1100
4080	16GB – 6x	9728	6.91	23 Gbps	2.51 Ghz	2200
3090	24 GB – 6	10496	7.45	19.5 Gbps	1.4 Ghz	1900
3090Ti	24 GB – 6	10752	7.64	21 Gbps	1.86 Ghz	2800

As you can see, the most balanced card is the RTX 4070 Super, which will be approximately 5× faster than my GTX 1660 based on the number of CUDA cores alone. Another more efficient and modern option is the RTX 5070 Ti, but its price is 30% greater, which isn’t worth it if we’re only talking about SD and have a limited budget.

Here is the list of the cards sorted by the Release Year and Generation

Card Name	VRAM	Release Year	NVIDIA Generation	Avg Price, eur
GTX 1660 Super	6GB	2019	Turing	250
RTX 3060 Ti	8GB	2020	Ampere	450
RTX 3080	12GB	2020	Ampere	1000
RTX 3090	24GB	2020	Ampere	1900
RTX 3060	12GB	2021	Ampere	300
RTX 3090 Ti	24GB	2022	Ampere	2800
RTX 4080	16GB	2022	Ada Lovelace	2200
RTX 4060 Ti	8GB	2023	Ada Lovelace	550
RTX 4070	12GB	2023	Ada Lovelace	650
RTX 4070 Ti	12GB	2023	Ada Lovelace	1200
RTX 4060 Ti	16GB	2023	Ada Lovelace	550
RTX 4070 Super	12GB	2024	Ada Lovelace	850
RTX 5070	12GB	2025	Blackwell	780
RTX 5070 Ti	16GB	2025	Blackwell	1100