PerGB

AMD Radeon RX 7900 XT

Live eBay prices, cheapest first. The best one right now is $850.00 (used).

Specs

VRAM
20 GB GDDR6
Bandwidth
800 GB/s
FP32
51.5 TFLOPS
FP16
103 TFLOPS
Power
315 W
Connector
PCIe 4.0
Released
2022
Launch price
$899.00
Llama 3.1 8B at Q4 (est.)
58–99 tokens/s

Price right now

Live on eBay
Lowest now$850.00
Median ask$999.00
Per GB VRAM$42.50
vs launch price5% under
Live listings29

From the live listings below; shipping included.

Listings

Live on eBay
ListingConditionPriceChecked
Used$850.00or best offer−15%shipping not listed
Open box$850.00or best offer−15%shipping not listed
Used$889.00or best offer−11%shipping not listed
Used$899.99−10%free shipping
Open box$900.00or best offer−10%free shipping
Used$912.24or best offer−9%free shipping
Used$945.00or best offer−5%free shipping
Refurbished$949.97free shipping
Used$949.99or best offershipping not listed
Refurbished$969.97free shipping
Used$970.00or best offerfree shipping
New$988.00free shipping
Refurbished$989.97free shipping
Used$990.97or best offerfree shipping
Used$999.00or best offershipping not listed
Used$1,000or best offershipping not listed
Used$1,000or best offershipping not listed
Used$1,000or best offershipping not listed
Used$1,018+$7.00 shipping
New$1,040free shipping
Used$1,044or best offerfree shipping
Used$1,099or best offerfree shipping
Used$1,119+$20.00 shipping
Used$1,150or best offerfree shipping
Used$1,150or best offerfree shipping
New$1,190+$20.00 shipping
New$1,500or best offershipping not listed
New$1,500free shipping
New$2,020+$20.00 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

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 RX 7900 XT neededCosts nowSpeed (est.)
Llama 3.1 8B1 card$850.0058–99 tokens/s
Qwen3 14B1 card$850.0031–53 tokens/s
gpt-oss 20B1 card$850.0077–132 tokens/s
Gemma 3 27B2 cards$1,70017–29 tokens/s
Qwen3 32B2 cards$1,70014–24 tokens/s
Llama 3.3 70B3 cards$2,5896.5–11 tokens/s
gpt-oss 120B4 cards$3,48954–93 tokens/s
Qwen3 235B-A22B8 cards$7,19613–22 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 RX 7900 XT neededCosts nowPer question (est.)
Julia-10.144B1 card$850.009.1–20 ms
Laya0.421B1 card$850.0015–33 ms
Kev-4B4B1 card$850.0036–78 ms
lev4B1 card$850.00109–234 ms
OpenJev27B1 card$850.00131–280 ms
Clef27B1 card$850.00131–280 ms
Perplexity Decider v1.227B1 card$850.00131–280 ms
mammoth-127B1 card$850.00131–280 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 RX 7900 XT. These open the shop's own search; we don't show their prices here.