How much memory does Qwen3-32B at Q4_K_M need?

23GB. That is 19GB of weights at 4.83 bits per weight, 2.1GB of K/V cache at 8K context, and 1.2GB of runtime overhead.

Which machine is cheapest for Qwen3-32B at Q4_K_M?

Mac mini, M6, 32GB, $1,499 at Apple. It fits with 6GB of headroom and decodes at an est. 5.7 tok/s.

Cheapest box to run Qwen3-32B at Q4_K_M

Alibaba’s Qwen3-32B is 32B parameters, dense. The number below is computed from that and each vendor’s published specs, not asserted. Switch the quant or the context and it recomputes in place.

Fit finderprices checked 43 minutes ago

Quant

Context

Qwen3-32B at Q4_K_M needs 23GB.

Cheapest Config that fits: Mac mini, M6, 32GB. Fits, 6GB headroom. Est. 5.7 tok/s.

Best offer today: $1,499 at Apple — in stock, checked 43 minutes ago.

Price history: 46 days recorded.

Offers — Mac mini, M6, 32GB, 512GB

Applelist pricein stockchecked 43 minutes ago$1,499
ExpercomSold out.out of stockchecked 41 days ago$1,499
B&H PhotoPreorder, ships about 22 Sep.out of stockchecked 45 days ago$1,599

Mac mini, M6, 16GB$1,099
Won't fit−10GB headroom
Mac mini, M6, 24GB$1,299
Won't fit−2GB headroom
Mac mini, M6, 32GB$1,499
Fits6GB headroomest. 5.7 tok/s
Mac mini, M5 Pro, 24GB$1,699
Won't fit−2GB headroom
Mac mini, M5 Pro, 48GB$2,499
Fits22GB headroomest. 10.3 tok/s
Mac Studio, M5 Max, 36GB$2,499
Fits10GB headroomest. 15.5 tok/s
Mac mini, M5 Pro, 64GB$2,899
Fits36GB headroomest. 10.3 tok/s
GeForce RTX 5090, 32GB$2,999.99+host
Fits8GB headroomest. 60.3 tok/s
Mac Studio, M4 Max, 64GB, refurbished$3,049
Fits36GB headroomest. 18.4 tok/s
Mac Studio, M5 Max, 48GB$3,099
Fits22GB headroomest. 20.7 tok/s
Mac Studio, M5 Max, 64GB$3,499
Fits36GB headroomest. 20.7 tok/s
Ascent GX10, GB10 Grace Blackwell, 128GB$3,999.99
Fits98GB headroomest. 9.2 tok/s
Mac Studio, M3 Ultra, 96GB, refurbished$4,329
Fits66GB headroomest. 27.6 tok/s
Mac Studio, M4 Max, 128GB, refurbished$4,409
Fits97GB headroomest. 18.4 tok/s
DGX Spark, GB10 Grace Blackwell, 128GB$4,699
Fits98GB headroomest. 9.2 tok/s
Mac Studio, M5 Max, 128GB$5,099
Fits97GB headroomest. 20.7 tok/s
Mac Studio, M5 Ultra, 96GB$5,499
Fits66GB headroomest. 40.4 tok/s
Mac Studio, M3 Ultra, 256GB, refurbished$7,729
Fits225GB headroomest. 27.6 tok/s
Mac Studio, M5 Ultra, 256GB$9,499
Fits225GB headroomest. 40.4 tok/s
Mac Studio, M5 Ultra, 512GB—
Fits481GB headroomest. 40.4 tok/s

Config	Memory	Price	Fit	Headroom	Est. speed
Mac mini, M6	16GB	$1,099	Won't fit	−10GB	—
Mac mini, M6	24GB	$1,299	Won't fit	−2GB	—
Mac mini, M6	32GB	$1,499	Fits	6GB	5.7 tok/s
Mac mini, M5 Pro	24GB	$1,699	Won't fit	−2GB	—
Mac mini, M5 Pro	48GB	$2,499	Fits	22GB	10.3 tok/s
Mac Studio, M5 Max	36GB	$2,499	Fits	10GB	15.5 tok/s
Mac mini, M5 Pro	64GB	$2,899	Fits	36GB	10.3 tok/s
GeForce RTX 5090	32GB	$2,999.99+host	Fits	8GB	60.3 tok/s
Mac Studio, M4 Max, refurb	64GB	$3,049	Fits	36GB	18.4 tok/s
Mac Studio, M5 Max	48GB	$3,099	Fits	22GB	20.7 tok/s
Mac Studio, M5 Max	64GB	$3,499	Fits	36GB	20.7 tok/s
Ascent GX10, GB10 Grace Blackwell	128GB	$3,999.99	Fits	98GB	9.2 tok/s
Mac Studio, M3 Ultra, refurb	96GB	$4,329	Fits	66GB	27.6 tok/s
Mac Studio, M4 Max, refurb	128GB	$4,409	Fits	97GB	18.4 tok/s
DGX Spark, GB10 Grace Blackwell	128GB	$4,699	Fits	98GB	9.2 tok/s
Mac Studio, M5 Max	128GB	$5,099	Fits	97GB	20.7 tok/s
Mac Studio, M5 Ultra	96GB	$5,499	Fits	66GB	40.4 tok/s
Mac Studio, M3 Ultra, refurb	256GB	$7,729	Fits	225GB	27.6 tok/s
Mac Studio, M5 Ultra	256GB	$9,499	Fits	225GB	40.4 tok/s
Mac Studio, M5 Ultra	512GB	—	Fits	481GB	40.4 tok/s

GeForce RTX 5090 is a card, not a computer: every price shown for it is the cheapest listing we found plus $1,000 for the minimum host machine.

memory = 32B × 0.604 B/param (4.83 bits) + 0.262 GB/1K × 8K K/V + 1.2 GB runtime
speed = bandwidth × 0.65 ÷ (32B active × 0.604 B/param) = 19.3 GB read per token
usable = memory − operating system reserve (8% on a Mac, 6% on a DGX Spark, 4% on a card)

Bandwidth is the vendor’s published figure. Every speed is an estimate and says so. Methodology.

What else Mac mini M6 32GB runs

Qwen3.5-35B-A3BQ5_K_M, est. 40.0 tok/s
Qwen3-32BQ5_K_M, est. 4.9 tok/s
Gemma-4-31BQ6_K, est. 4.4 tok/s
Qwen3-30B-A3BQ6_K, est. 31.4 tok/s
Gemma-3-27BQ6_K, est. 5.0 tok/s
Qwen3.6-27BQ6_K, est. 5.0 tok/s

Best quant that still fits at 8K context, largest model first. Full list on the Config page.

Bandwidth 170GB/s, usable memory 29GB of 32GB.