buyaigear
Prices checked 6 min ago

NVIDIA · Ada Lovelace · 24 GB · Qwen3.8 27B · Q4_K_M: ~40 tok/s

NVIDIA GeForce RTX 4090

Still the reference card for local AI: 24 GB at 1 TB/s, which puts a 32B model at Q4 in the mid 30s of tokens per second. Used prices track the 5090 closely.

Lowest price right now

$2,083.48

eBay · $2,073.49 + $9.99 shipping · $86 per GB · observed 2026-10-14 22:18 UTC

See it on eBaySee it on Amazon

Where to buy

VendorPriceConditionObserved
eBay
cheapest active listing		 $2,083.48
$2,073.49 + $9.99 shipping		 new and used 2026-10-14 22:18 UTC Go to eBay
Amazon
price tracking coming		 price on Amazon new Go to Amazon

On eBay now

378 listings passed our filters at the last check. Cheapest delivered first. Seller reputation is feedback count and positive rate.

All 378 listings · Open the search on eBay

What it runs

For each model, grouped by vendor with the newest first: the largest quant that fits with 8K of context and the estimated speed for one user. The model you picked on the ladder is highlighted. Estimates from the spec sheet, not measurements: see how we estimate.

ModelQuantNeeds at 8KFitMax contextDecode tok/sPrefill tok/s
DeepSeek
DeepSeek V4.1 Flash 552B-A16B Sep 2026Q4_K_M338 GBdoes not fit
DeepSeek V4 Flash 284B-A13B Jul 2026Q4_K_M175 GBdoes not fit
DeepSeek V3.1 671B-A37B Aug 2025Q4_K_M411 GBdoes not fit
Gemma
Gemma 4 12B May 2026Q8_014 GBfits128K+~52~2,400
Gemma 4 31B Mar 2026Q5_K_M24 GBfits8K~30~940
Gemma 4 26B-A4B Mar 2026Q6_K22 GBfits87K~210~7,600
Gemma 3 27B Mar 2025Q5_K_M22 GBfits36K~34~1,100
GLM
GLM-5.3 753B-A40B Aug 2026Q4_K_M462 GBdoes not fit
GLM-5.3 Flash 320B-A18B Aug 2026Q4_K_M197 GBdoes not fit
GLM-4.6 355B-A32B Sep 2025Q4_K_M221 GBdoes not fit
GLM-4.5 Air 106B-A12B Jul 2025Q4_K_M68 GBdoes not fit
gpt-oss
gpt-oss-120b Aug 2025Q4_K_M73 GBdoes not fit
gpt-oss-20b Aug 2025Q6_K19 GBfits104K~220~8,000
Llama
Llama 4 Scout 109B-A17B Apr 2025Q4_K_M70 GBdoes not fit
Llama 3.3 70B Dec 2024Q4_K_M47 GBdoes not fit
Llama 3.1 8B Jul 2024Q8_011 GBfits104K~77~3,600
MiMo
MiMo V2.6 Pro 1.02T-A42B Sep 2026Q4_K_M624 GBdoes not fit
MiMo V2.6 Flash 309B-A15B Sep 2026Q4_K_M190 GBdoes not fit
Mistral
Mistral Medium 3.5 128B Mar 2026Q4_K_M83 GBdoes not fit
Mistral Small 4 119B-A6.5B Mar 2026Q4_K_M79 GBdoes not fit
Mistral Small 3.2 24B Jun 2025Q6_K23 GBfits16K~33~1,200
Qwen
Qwen3.8 Flash-Next 125B-A6B Aug 2026Q4_K_M78 GBdoes not fit
Qwen3.8 27B Aug 2026 (your pick)Q5_K_M21 GBfits49K~34~1,100
Qwen3.6 27B Apr 2026Q5_K_M21 GBfits49K~34~1,100
Qwen3.6 35B-A3B Apr 2026Q4_K_M23 GBfits54K~360~9,600
Qwen3 Coder 480B-A35B Jul 2025Q4_K_M296 GBdoes not fit
Qwen3 235B-A22B Jul 2025Q4_K_M146 GBdoes not fit
Qwen3 32B Apr 2025Q4_K_M24 GBfits9K~33~880

Specifications

Memory
24 GB GDDR6X
Memory bandwidth
1,008 GB/s
Bus width
384-bit
Shader cores
16,384
Tensor cores
512
FP32
82.6 TFLOPS
FP16 (dense)
165 TFLOPS
FP8 (dense)
330 TFLOPS
Power
450 W
Form factor
Triple slot, full height
Interface
PCIe 4.0 x16
Architecture
Ada Lovelace
Launched
2022
MSRP
$1,599

Source: manufacturer specifications.