PerGB

NVIDIA Titan RTX

Live eBay prices, cheapest first. The best one right now is $910.00 (used).

Specs

VRAM
24 GB GDDR6
Bandwidth
672 GB/s
FP32
16.3 TFLOPS
FP16
130 TFLOPS
Power
280 W
Connector
PCIe 3.0
Released
2018
Launch price
$2,499
CUDA compute
7.5
Llama 3.1 8B at Q4 (est.)
69–104 tokens/s

Price right now

Live on eBay
Lowest now$910.00
Median ask$1,332
Per GB VRAM$37.92
vs launch price64% under
Live listings54

From the live listings below; shipping included.

Listings

Live on eBay
ListingConditionPriceChecked
Used$910.00−32%Check carefullyfree shipping
Used$929.99−30%free shipping
Used$1,000−25%free shipping
Used$1,000or best offer−25%free shipping
Used$1,019or best offer−23%free shipping
Used$1,020−23%+$70.00 shipping
Used$1,020−23%free shipping
Used$1,098or best offer−18%free shipping
Used$1,099or best offer−17%free shipping
Used$1,100or best offer−17%free shipping
Used$1,100or best offer−17%free shipping
Used$1,100or best offer−17%shipping not listed
Used$1,100or best offer−17%free shipping
Used$1,100−17%+$20.00 shipping
Used$1,100or best offer−17%free shipping
Used$1,100or best offer−17%free shipping
Used$1,138or best offer−15%+$19.99 shipping
Used$1,165or best offer−13%+$15.00 shipping
Used$1,169or best offer−12%free shipping
Used$1,225or best offer−8%free shipping
Used$1,246or best offer−6%free shipping
Used$1,249−6%free shipping
Used$1,280+$50.00 shipping
Used$1,304or best offerfree shipping
Used$1,311or best offerfree shipping
Used$1,319or best offerfree shipping
Used$1,322or best offerfree shipping
Used$1,342or best offerfree shipping
Used$1,348free shipping
Used$1,348free shipping
Used$1,348free shipping
Used$1,348free shipping
Used$1,349or best offerfree shipping
Used$1,350free shipping
Used$1,350or best offerfree shipping
Used$1,353+$5.00 shipping
Used$1,354+$5.00 shipping
Used$1,355or best offerfree shipping
Used$1,355or best offerfree shipping
Used$1,355or best offerfree shipping
Refurbished$1,390free shipping
Used$1,455or best offerfree shipping
Used$1,455or best offerfree shipping
Used$1,508+$38.88 shipping
Used$1,508+$38.88 shipping
Used$1,508+$38.88 shipping
Used$1,511or best offer+$38.88 shipping
Used$1,568or best offer+$38.88 shipping
Used$1,585+$38.80 shipping
Used$1,599free shipping
Used$1,600or best offerfree shipping
Open box$1,604+$10.00 shipping
Open box$2,080+$80.00 shipping
New$3,298or best offer+$10.00 shipping

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 Quadro RTX 6000, vs Quadro RTX 8000, vs RTX A5000.

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)Titan RTX neededCosts nowSpeed (est.)
Llama 3.1 8B1 card$910.0069–104 tokens/s
Qwen3 14B1 card$910.0037–56 tokens/s
gpt-oss 20B1 card$910.0092–139 tokens/s
Gemma 3 27B1 card$910.0020–30 tokens/s
Qwen3 32B1 card$910.0017–25 tokens/s
Llama 3.3 70B2 cards$1,8407.9–12 tokens/s
gpt-oss 120B4 cards$3,84065–98 tokens/s
Qwen3 235B-A22B7 cards$6,89915–23 tokens/s
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)Titan RTX neededCosts nowPer question (est.)
Julia-10.144B1 card$910.006.7–14 ms
Laya0.421B1 card$910.0011–24 ms
Kev-4B4B1 card$910.0027–58 ms
lev4B1 card$910.0081–173 ms
OpenJev27B1 card$910.0096–207 ms
Clef27B1 card$910.0096–207 ms
Perplexity Decider v1.227B1 card$910.0096–207 ms
mammoth-127B1 card$910.0096–207 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 Titan RTX. These open the shop's own search; we don't show their prices here.