PerGB

NVIDIA GeForce RTX 4070 Super

Live eBay prices, cheapest first. The best one right now is $700.00 (used).

Specs

VRAM
12 GB GDDR6X
Bandwidth
504 GB/s
FP32
35.5 TFLOPS
FP16
71 TFLOPS
Power
220 W
Connector
PCIe 4.0
Released
2024
Launch price
$599.00
CUDA compute
8.9
Llama 3.1 8B at Q4 (est.)
52–78 tokens/s

Price right now

Live on eBay
Lowest now$700.00
Median ask$1,100
Per GB VRAM$58.33
vs launch price17% over
Live listings90

From the live listings below; shipping included.

Listings

Live on eBay
ListingConditionPriceChecked
Used$700.00or best offer−36%Check carefullyshipping not listed
Used$724.97−34%free shipping
Used$724.99or best offer−34%free shipping
Used$729.99−34%free shipping
Used$735.00or best offer−33%shipping not listed
Used$749.00or best offer−32%Check carefullyshipping not listed
Used$749.99−32%free shipping
Used$759.99−31%free shipping
Used$780.00or best offer−29%shipping not listed
Used$786.31or best offer−29%+$16.31 shipping
New$799.00or best offer−27%free shipping
Refurbished$799.99−27%free shipping
Open box$809.99−26%free shipping
New$820.00or best offer−25%shipping not listed
Used$830.00or best offer−25%+$30.00 shipping
Refurbished$849.94−23%free shipping
Open box$850.00−23%shipping not listed
Used$850.00or best offer−23%shipping not listed
Used$850.00or best offer−23%shipping not listed
Open box$854.98−22%+$9.99 shipping
Used$875.26or best offer−20%shipping not listed
Used$875.26or best offer−20%shipping not listed
Refurbished$899.99−18%free shipping
Open box$907.94or best offer−17%+$7.94 shipping
New$920.63or best offer−16%free shipping
Used$929.99−15%free shipping
Used$950.00or best offer−14%shipping not listed
Used$960.00−13%free shipping
New$967.31or best offer−12%+$5.99 shipping
New$970.54or best offer−12%+$5.99 shipping
New$980.09or best offer−11%+$5.99 shipping
New$980.73−11%+$5.99 shipping
Used$981.08−11%shipping not listed
New$998.53or best offer−9%+$5.99 shipping
New$999.00or best offer−9%free shipping
New$1,000or best offer−9%shipping not listed
New$1,002−9%+$5.99 shipping
Used$1,002or best offer−9%+$11.02 shipping
New$1,006or best offer−9%+$5.99 shipping
Used$1,008−8%+$20.00 shipping
Used$1,049or best offerfree shipping
Used$1,050or best offerfree shipping
Used$1,095or best offerfree shipping
New$1,099free shipping
Used$1,100or best offerfree shipping
Refurbished$1,100free shipping
Refurbished$1,100free shipping
Refurbished$1,100free shipping
Refurbished$1,100free shipping
Refurbished$1,100free shipping
Refurbished$1,100free shipping
Refurbished$1,100free shipping
Refurbished$1,100free shipping
Used$1,112or best offer+$11.02 shipping
Used$1,118or best offer+$16.53 shipping
Used$1,135or best offerfree shipping
New$1,148or best offerfree shipping
Used$1,150or best offerfree shipping
Used$1,150or best offerfree shipping
Used$1,151or best offerfree shipping

Showing the 60 cheapest of 90 listings.

Auctions ending soon

No auctions ending soon right now.

Prices include shipping and were read from eBay at the time shown. They can change: check the listing before you buy.

Owner reviews

No owner reviews yet. Own one? Tell others how it runs.

From signed-in owners, checked before they're published. PerGB doesn't pay for or edit reviews.

How it compares

Live on eBay

As much VRAM, about as fast, for less

Faster, for the least

More VRAM, for not much more

Compare: vs L4, vs RTX A4000, vs GeForce RTX 2080 Ti 11 GB.

From the live listings of every GPU, shipping included; performance against the GeForce RTX 3090 (FP16 or FP32 compute and memory bandwidth). Updated with every build.

Local AI: what it runs

Model (Q4, 8K context)GeForce RTX 4070 Super neededCosts nowSpeed (est.)
Llama 3.1 8B1 card$700.0052–78 tokens/s
Qwen3 14B1 card$700.0028–42 tokens/s
gpt-oss 20B2 cards$1,42569–104 tokens/s
Gemma 3 27B2 cards$1,42515–23 tokens/s
Qwen3 32B3 cards$2,15013–19 tokens/s
Llama 3.3 70B5 cards$3,6155.9–8.8 tokens/s
gpt-oss 120B7 cards$5,11449–73 tokens/s
Qwen3 235B-A22Bmore than 8——
DeepSeek R1 / V3 671Bmore than 8——

Cost: that many of the cheapest live listings. Speed: an estimate from the memory bandwidth for one person chatting. Try other models, sizes and contexts.

Decision models

Model (Q4, 8K context)GeForce RTX 4070 Super neededCosts nowPer question (est.)
Julia-10.144B1 card$700.0011–23 ms
Laya0.421B1 card$700.0018–38 ms
Kev-4B4B1 card$700.0042–90 ms
lev4B1 card$700.00126–270 ms
OpenJev27B2 cards$1,425151–323 ms
Clef27B2 cards$1,425151–323 ms
Perplexity Decider v1.227B2 cards$1,425151–323 ms
mammoth-127B2 cards$1,425151–323 ms

Decision models score the options you give them instead of writing text, so they're timed per question. The time is scaled from each model's reported speed on an RTX PRO 6000 by this card's compute and memory bandwidth, within about half either way. About the models.

Compare elsewhere

Search other shops for NVIDIA GeForce RTX 4070 Super. These open the shop's own search; we don't show their prices here.