PerGB

NVIDIA RTX A4000

Live eBay prices, cheapest first. The best one right now is $949.99 (used).

Specs

VRAM
16 GB GDDR6
Bandwidth
448 GB/s
FP32
19.2 TFLOPS
FP16
77 TFLOPS
Power
140 W
Connector
PCIe 4.0
Released
2021
Launch price
$1,000
CUDA compute
8.6
Llama 3.1 8B at Q4 (est.)
46–69 tokens/s

Price right now

Live on eBay
Lowest now$949.99
Median ask$1,377
Per GB VRAM$59.37
vs launch price5% under
Live listings90

From the live listings below; shipping included.

Listings

Live on eBay
ListingConditionPriceChecked
Used$949.99−31%free shipping
Used$960.00−30%shipping not listed
Used$960.00or best offer−30%shipping not listed
Used$964.00−30%+$25.00 shipping
Refurbished$969.95−30%free shipping
Refurbished$969.99−30%free shipping
Refurbished$969.99−30%free shipping
Used$974.99or best offer−29%+$25.00 shipping
Used$979.97−29%free shipping
Used$979.99−29%free shipping
Used$995.00−28%shipping not listed
Open box$999.00−27%free shipping
Used$999.00or best offer−27%free shipping
Open box$999.95−27%free shipping
Used$999.99or best offer−27%free shipping
Used$999.99−27%free shipping
Used$999.99−27%free shipping
Used$1,000or best offer−27%shipping not listed
Used$1,000or best offer−27%shipping not listed
Used$1,016−26%shipping not listed
Used$1,016or best offer−26%shipping not listed
Used$1,016or best offer−26%shipping not listed
Used$1,020or best offer−26%+$20.00 shipping
Refurbished$1,050−24%free shipping
Used$1,050or best offer−24%shipping not listed
Used$1,090or best offer−21%free shipping
New$1,100or best offer−20%shipping not listed
Used$1,125or best offer−18%+$24.99 shipping
Open box$1,130−18%free shipping
Used$1,144or best offer−17%free shipping
Used$1,145or best offer−17%free shipping
Used$1,146or best offer−17%free shipping
New$1,150or best offer−16%shipping not listed
Used$1,184or best offer−14%free shipping
Used$1,198or best offer−13%free shipping
Used$1,199−13%shipping not listed
Used$1,199or best offer−13%+$100.00 shipping
Refurbished$1,200−13%free shipping
New$1,200or best offer−13%shipping not listed
New$1,230−11%shipping not listed
Used$1,274or best offer−7%free shipping
Used$1,299or best offer−6%free shipping
Open box$1,299−6%free shipping
Used$1,300or best offer−6%free shipping
Used$1,377+$275.58 shipping
Used$1,377+$275.58 shipping
Used$1,377+$275.58 shipping
Used$1,395or best offerfree shipping
Open box$1,445+$5.00 shipping
Used$1,498or best offerfree shipping
Used$1,498or best offerfree shipping
Used$1,498or best offerfree shipping
Used$1,498or best offerfree shipping
New$1,510or best offerfree shipping
Refurbished$1,511+$98.10 shipping
Refurbished$1,511+$98.10 shipping
New$1,520or best offer+$20.00 shipping
Open box$1,597+$275.58 shipping
Used$1,599or best offerfree shipping
Open box$1,599or best offershipping not listed

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 GeForce RTX 2080 Ti 11 GB, vs GeForce RTX 2080 Ti 22 GB, vs GeForce RTX 4070 Super, vs Radeon RX 9060 XT 16 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)RTX A4000 neededCosts nowSpeed (est.)
Llama 3.1 8B1 card$949.9946–69 tokens/s
Qwen3 14B1 card$949.9925–37 tokens/s
gpt-oss 20B1 card$949.9962–92 tokens/s
Gemma 3 27B2 cards$1,91013–20 tokens/s
Qwen3 32B2 cards$1,91011–17 tokens/s
Llama 3.3 70B4 cards$3,8345.2–7.9 tokens/s
gpt-oss 120B5 cards$4,80443–65 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)RTX A4000 neededCosts nowPer question (est.)
Julia-10.144B1 card$949.9911–23 ms
Laya0.421B1 card$949.9918–38 ms
Kev-4B4B1 card$949.9943–92 ms
lev4B1 card$949.99128–275 ms
OpenJev27B2 cards$1,910153–329 ms
Clef27B2 cards$1,910153–329 ms
Perplexity Decider v1.227B2 cards$1,910153–329 ms
mammoth-127B2 cards$1,910153–329 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 RTX A4000. These open the shop's own search; we don't show their prices here.