PerGB

NVIDIA A10

Live eBay prices, cheapest first. The best one right now is $1,930 (used).

Specs

VRAM
24 GB GDDR6
Bandwidth
600 GB/s
FP32
31.2 TFLOPS
FP16
125 TFLOPS
Power
150 W
Connector
PCIe 4.0
Released
2021
Launch price
$3,000 (est.)
CUDA compute
8.6
Llama 3.1 8B at Q4 (est.)
62–93 tokens/s

Running it at home

Cooling
Passive: it has no fan. In a desktop it needs air pushed through it, front to back: a fan shroud on its end (3D-printed ones with a 40 mm blower or a 92 mm fan are sold for most of these cards) or a server chassis. Without one it overheats within minutes under load.
Power
One 8-pin CPU (EPS-12V) power input, not a PCIe one: it needs an adapter cable (two PCIe 8-pin to one EPS 8-pin, sold for these cards). A PCIe plug fits but is wired differently: never force one in. Rated 150 W: allow that for each card on top of the rest of the PC.
Size
Full height, two slots.
Motherboard
Turn on "Above 4G decoding" in the BIOS: without it many boards won't start with a card with this much memory.
Screen
No display outputs: use the processor's graphics or a small second card for a monitor.

A server card: made for a rack server's airflow and power. With these sorted it works in a normal PC.

Price right now

Live on eBay
Lowest now$1,930
Median ask$4,688
Per GB VRAM$80.42
vs launch price36% under
Live listings45

From the live listings below; shipping included.

Listings

Live on eBay
ListingConditionPriceChecked
Used$1,930or best offerCheck carefully+$30.00 shipping
Used$4,019−14%+$20.00 shipping
Open box$4,188or best offer−11%free shipping
Open box$4,188or best offer−11%free shipping
New$4,290or best offer−8%free shipping
Open box$4,396or best offer−6%free shipping
Open box$4,399or best offer−6%free shipping
New$4,625or best offerfree shipping
New$4,626or best offerfree shipping
New$4,648+$230.00 shipping
Open box$4,649or best offerfree shipping
Open box$4,649or best offerfree shipping
New$4,650+$232.00 shipping
New$4,650or best offerfree shipping
New$4,650or best offerfree shipping
Open box$4,650or best offerfree shipping
Open box$4,650or best offerfree shipping
Open box$4,650or best offerfree shipping
Open box$4,650or best offerfree shipping
Open box$4,650or best offerfree shipping
New$4,688or best offerfree shipping
New$4,688or best offerfree shipping
New$4,688or best offerfree shipping
New$4,688or best offerfree shipping
New$4,688or best offerfree shipping
New$4,693or best offer+$4.99 shipping
New$4,693or best offer+$4.99 shipping
New$4,693or best offer+$4.99 shipping
New$4,693or best offer+$4.99 shipping
New$4,750or best offerfree shipping
New$4,750or best offerfree shipping
New$4,750or best offerfree shipping
New$4,879+$229.00 shipping
Open box$4,881+$231.00 shipping
Open box$4,885or best offerfree shipping
New$5,015or best offer+$38.88 shipping
Open box$5,245or best offerfree shipping
Open box$5,245or best offerfree shipping
Open box$5,245or best offerfree shipping
Open box$5,245or best offerfree shipping
Open box$5,245or best offerfree shipping
Open box$5,245or best offerfree shipping
New$5,371or best offerfree shipping
Open box$5,489or best offerfree shipping
New$5,550or best offerfree 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

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)A10 neededCosts nowSpeed (est.)
Llama 3.1 8B1 card$1,93062–93 tokens/s
Qwen3 14B1 card$1,93033–50 tokens/s
gpt-oss 20B1 card$1,93082–124 tokens/s
Gemma 3 27B1 card$1,93018–27 tokens/s
Qwen3 32B1 card$1,93015–23 tokens/s
Llama 3.3 70B2 cards$5,9497.0–11 tokens/s
gpt-oss 120B4 cards$14,32458–87 tokens/s
Qwen3 235B-A22B7 cards$27,40913–20 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)A10 neededCosts nowPer question (est.)
Julia-10.144B1 card$1,9307.3–16 ms
Laya0.421B1 card$1,93012–26 ms
Kev-4B4B1 card$1,93029–62 ms
lev4B1 card$1,93087–187 ms
OpenJev27B1 card$1,930104–223 ms
Clef27B1 card$1,930104–223 ms
Perplexity Decider v1.227B1 card$1,930104–223 ms
mammoth-127B1 card$1,930104–223 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 A10. These open the shop's own search; we don't show their prices here.