PerGB

AMD Radeon Pro W7900

Live eBay prices, cheapest first. The best one right now is $3,630 (used).

Specs

VRAM
48 GB GDDR6
Bandwidth
864 GB/s
FP32
61.3 TFLOPS
FP16
123 TFLOPS
Power
295 W
Connector
PCIe 4.0
Released
2023
Launch price
$3,999
Llama 3.1 8B at Q4 (est.)
62–107 tokens/s

Price right now

Live on eBay
Lowest now$3,630
Median ask$5,099
Per GB VRAM$75.62
vs launch price9% under
Live listings13

From the live listings below; shipping included.

Listings

Live on eBay
ListingConditionPriceChecked
Used$3,630or best offer−29%+$126.14 shipping
Refurbished$4,000−22%free shipping
New$4,100−20%free shipping
New$5,069or best offerfree shipping
New$5,099or best offerfree shipping
New$5,099or best offerfree shipping
New$5,099or best offerfree shipping
New$5,500or best offerfree shipping
New$6,315or best offerfree shipping
New$6,315or best offerfree shipping
New$6,353+$52.80 shipping
New$6,500or best offerfree shipping
New$8,954or best offer+$38.88 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)Radeon Pro W7900 neededCosts nowSpeed (est.)
Llama 3.1 8B1 card$3,63062–107 tokens/s
Qwen3 14B1 card$3,63034–58 tokens/s
gpt-oss 20B1 card$3,63083–143 tokens/s
Gemma 3 27B1 card$3,63018–31 tokens/s
Qwen3 32B1 card$3,63015–26 tokens/s
Llama 3.3 70B1 card$3,6307.1–12 tokens/s
gpt-oss 120B2 cards$7,63059–101 tokens/s
Qwen3 235B-A22B4 cards$16,79914–23 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)Radeon Pro W7900 neededCosts nowPer question (est.)
Julia-10.144B1 card$3,6308.0–17 ms
Laya0.421B1 card$3,63013–29 ms
Kev-4B4B1 card$3,63032–69 ms
lev4B1 card$3,63096–206 ms
OpenJev27B1 card$3,630115–246 ms
Clef27B1 card$3,630115–246 ms
Perplexity Decider v1.227B1 card$3,630115–246 ms
mammoth-127B1 card$3,630115–246 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 AMD Radeon Pro W7900. These open the shop's own search; we don't show their prices here.