PerGB

NVIDIA A100 SXM4 64 GB

Live eBay prices, cheapest first. The best one right now is $9,335 (open box).

Not a PCIe card. SXM module: needs an SXM-to-PCIe adapter board, its own cooling and power to run in a normal PC. The PCIe version: NVIDIA A100 PCIe 40 GB, from $5,588 →

Other sizes: 40 GB, 80 GB

Specs

VRAM
64 GB HBM2e
Bandwidth
2,039 GB/s
FP32
19.5 TFLOPS
FP16
312 TFLOPS
Power
400 W
Connector
SXM4 (PCIe 4.0)
Released
2020
CUDA compute
8.0
Llama 3.1 8B at Q4 (est.)
210–315 tokens/s

Price right now

Live on eBay
Lowest now$9,335
Median ask$9,450
Per GB VRAM$145.86
vs launch price—
Live listings11

From the live listings below; shipping included.

Listings

Live on eBay
ListingConditionPriceChecked
Open box$9,335or best offerfree shipping
Open box$9,450or best offerfree shipping
Open box$9,450or best offerfree shipping
Open box$9,450or best offerfree shipping
Open box$9,450or best offerfree shipping
Open box$9,450or best offerfree shipping
Open box$9,450or best offerfree shipping
Open box$9,450or best offerfree shipping
Open box$9,780or best offerfree shipping
Open box$11,568or best offerfree shipping
Open box$11,568or 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)A100 SXM4 64 GB neededCosts nowSpeed (est.)
Llama 3.1 8B1 card$9,335210–315 tokens/s
Qwen3 14B1 card$9,335114–170 tokens/s
gpt-oss 20B1 card$9,335280–420 tokens/s
Gemma 3 27B1 card$9,33561–92 tokens/s
Qwen3 32B1 card$9,33551–77 tokens/s
Llama 3.3 70B1 card$9,33524–36 tokens/s
gpt-oss 120B2 cards$18,785198–297 tokens/s
Qwen3 235B-A22B3 cards$28,23546–69 tokens/s
DeepSeek R1 / V3 671B7 cards$66,03527–41 tokens/s

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)A100 SXM4 64 GB neededCosts nowPer question (est.)
Julia-10.144B1 card$9,3352.5–5.3 ms
Laya0.421B1 card$9,3354.2–8.9 ms
Kev-4B4B1 card$9,33510.0–21 ms
lev4B1 card$9,33530–64 ms
OpenJev27B1 card$9,33536–77 ms
Clef27B1 card$9,33536–77 ms
Perplexity Decider v1.227B1 card$9,33536–77 ms
mammoth-127B1 card$9,33536–77 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 A100 SXM4 64 GB. These open the shop's own search; we don't show their prices here.