PerGB

NVIDIA GeForce RTX 4070 Ti Super

Live eBay prices, cheapest first. The best one right now is $999.99 (used).

Specs

VRAM
16 GB GDDR6X
Bandwidth
672 GB/s
FP32
44.1 TFLOPS
FP16
88 TFLOPS
Power
285 W
Connector
PCIe 4.0
Released
2024
Launch price
$799.00
CUDA compute
8.9
Llama 3.1 8B at Q4 (est.)
69–104 tokens/s

Price right now

Live on eBay
Lowest now$999.99
Median ask$1,606
Per GB VRAM$62.50
vs launch price25% over
Live listings126

From the live listings below; shipping included.

Listings

Live on eBay
ListingConditionPriceChecked
Used$999.99or best offer−38%Check carefullyfree shipping
Used$1,000or best offer−38%shipping not listed
Used$1,050−35%Check carefullyshipping not listed
Open box$1,100−32%+$9.99 shipping
Open box$1,100or best offer−32%shipping not listed
Used$1,100or best offer−32%shipping not listed
Used$1,100or best offer−32%shipping not listed
Used$1,100−32%Check carefullyshipping not listed
Used$1,100or best offer−32%shipping not listed
Open box$1,135or best offer−29%+$85.00 shipping
Used$1,160−28%free shipping
Used$1,160−28%free shipping
Used$1,160−28%free shipping
Used$1,200−25%free shipping
Open box$1,200or best offer−25%shipping not listed
Refurbished$1,200−25%free shipping
Used$1,200or best offer−25%shipping not listed
Open box$1,239or best offer−23%+$17.00 shipping
Used$1,250or best offer−22%shipping not listed
Used$1,250or best offer−22%shipping not listed
Open box$1,272−21%+$121.22 shipping
Used$1,300−19%free shipping
Used$1,300−19%free shipping
Used$1,301−19%+$2.97 shipping
New$1,371or best offer−15%+$1.65 shipping
Open box$1,390or best offer−13%free shipping
New$1,393or best offer−13%free shipping
Open box$1,400or best offer−13%free shipping
Used$1,400or best offer−13%shipping not listed
Used$1,400or best offer−13%shipping not listed
Used$1,402−13%shipping not listed
Used$1,403or best offer−13%+$24.79 shipping
Used$1,409−12%+$20.00 shipping
Open box$1,417or best offer−12%free shipping
New$1,450−10%free shipping
Refurbished$1,479−8%free shipping
New$1,480or best offer−8%free shipping
Used$1,489−7%+$20.00 shipping
Used$1,493−7%+$201.00 shipping
New$1,500or best offer−7%shipping not listed
Used$1,520−5%+$20.00 shipping
New$1,531+$450.00 shipping
Open box$1,549or best offerfree shipping
Open box$1,549or best offerfree shipping
Open box$1,549or best offerfree shipping
New$1,562or best offer+$29.90 shipping
New$1,596or best offerfree shipping
Used$1,598or best offerfree shipping
Used$1,599or best offerfree shipping
Refurbished$1,600free shipping
Refurbished$1,600free shipping
Refurbished$1,600free shipping
Refurbished$1,600free shipping
Refurbished$1,600free shipping
Refurbished$1,600free shipping
Refurbished$1,600free shipping
Refurbished$1,600free shipping
Refurbished$1,600free shipping
Refurbished$1,600free shipping
Refurbished$1,600free shipping

Showing the 60 cheapest of 126 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 A16, vs GeForce RTX 3080 Ti, vs GeForce RTX 3090.

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 Ti Super neededCosts nowSpeed (est.)
Llama 3.1 8B1 card$999.9969–104 tokens/s
Qwen3 14B1 card$999.9937–56 tokens/s
gpt-oss 20B1 card$999.9992–139 tokens/s
Gemma 3 27B2 cards$2,00020–30 tokens/s
Qwen3 32B2 cards$2,00017–25 tokens/s
Llama 3.3 70B4 cards$4,1507.9–12 tokens/s
gpt-oss 120B5 cards$5,25065–98 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 Ti Super neededCosts nowPer question (est.)
Julia-10.144B1 card$999.998.2–18 ms
Laya0.421B1 card$999.9914–29 ms
Kev-4B4B1 card$999.9933–70 ms
lev4B1 card$999.9998–210 ms
OpenJev27B2 cards$2,000117–251 ms
Clef27B2 cards$2,000117–251 ms
Perplexity Decider v1.227B2 cards$2,000117–251 ms
mammoth-127B2 cards$2,000117–251 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 Ti Super. These open the shop's own search; we don't show their prices here.