PerGB

Intel Arc A770 8 GB

Live eBay prices, cheapest first. The best one right now is $259.99 (used).

Other sizes: 16 GB

Specs

VRAM
8 GB GDDR6
Bandwidth
512 GB/s
FP32
19.7 TFLOPS
FP16
138 TFLOPS
Power
225 W
Connector
PCIe 4.0
Released
2022
Launch price
$329.00
Llama 3.1 8B at Q4 (est.)
37–63 tokens/s

Price right now

Live on eBay
Lowest now$259.99
Median ask$530.07
Per GB VRAM$32.50
vs launch price21% under
Live listings17

From the live listings below; shipping included.

Listings

Live on eBay
ListingConditionPriceChecked
Used$259.99free shipping
Used$265.00or best offershipping not listed
Open box$280.00or best offerCheck carefullyfree shipping
Used$299.00or best offer−44%free shipping
Used$299.99−43%free shipping
Used$340.00−36%+$20.00 shipping
Used$419.99−21%free shipping
New$510.49or best offerfree shipping
Open box$530.07+$60.00 shipping
New$530.07+$60.00 shipping
New$536.49or best offerfree shipping
New$539.49or best offerfree shipping
Open box$539.49or best offerfree shipping
New$539.49or best offerfree shipping
New$554.49+$60.00 shipping
New$554.49+$60.00 shipping
New$559.98+$60.00 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)Arc A770 8 GB neededCosts nowSpeed (est.)
Llama 3.1 8B1 card$259.9937–63 tokens/s
Qwen3 14B2 cards$524.9920–34 tokens/s
gpt-oss 20B2 cards$524.9949–84 tokens/s
Gemma 3 27B4 cards$1,10411–18 tokens/s
Qwen3 32B4 cards$1,1049.0–15 tokens/s
Llama 3.3 70B7 cards$2,1644.2–7.2 tokens/s
gpt-oss 120Bmore than 8——
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)Arc A770 8 GB neededCosts nowPer question (est.)
Julia-10.144B1 card$259.999.8–21 ms
Laya0.421B1 card$259.9916–35 ms
Kev-4B4B1 card$259.9939–84 ms
lev4B1 card$259.99118–253 ms
OpenJev27B3 cards$804.99141–302 ms
Clef27B3 cards$804.99141–302 ms
Perplexity Decider v1.227B3 cards$804.99141–302 ms
mammoth-127B3 cards$804.99141–302 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 Intel Arc A770 8 GB. These open the shop's own search; we don't show their prices here.