How much memory does Gemma-3-27B at Q4_K_M need?

20GB. That is 16GB of weights at 4.83 bits per weight, 2.0GB of K/V cache at 8K context, and 1.2GB of runtime overhead.

Which machine is cheapest for Gemma-3-27B at Q4_K_M?

Mac mini, M6, 32GB, $1,499 at Apple. It fits with 9GB of headroom and decodes at an est. 6.8 tok/s.

Cheapest box to run Gemma-3-27B at Q4_K_M

Google’s Gemma-3-27B is 27B 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 10 hours ago

Quant

Context

Gemma-3-27B at Q4_K_M needs 20GB.

Cheapest Config that fits: Mac mini, M6, 32GB. Fits, 9GB headroom. Est. 6.8 tok/s.

Best offer today: $1,499 at Apple — in stock, checked 10 hours ago.

Collecting price history — 2 of 90 days. No timing verdict yet.

Offers — Mac mini, M6, 32GB, 512GB

Expercomout of stockchecked 10 hours ago$1,499
Applelist pricein stockchecked 10 hours ago$1,499
B&H Photoout of stockchecked 10 hours ago$1,599

Mac mini, M6, 16GB$1,099
Won't fit−7GB headroom
Mac mini, M6, 24GB$1,299
Tight1GB headroomest. 6.8 tok/s
Mac mini, M6, 32GB$1,499
Fits9GB headroomest. 6.8 tok/s
Mac mini, M5 Pro, 24GB$1,699
Tight1GB headroomest. 12.2 tok/s
Mac mini, M5 Pro, 48GB$2,499
Fits25GB headroomest. 12.2 tok/s
Mac Studio, M5 Max, 36GB$2,499
Fits13GB headroomest. 24.5 tok/s
Mac mini, M5 Pro, 64GB$2,899
Fits39GB headroomest. 12.2 tok/s
GeForce RTX 5090, 32GB$2,999.99+host
Fits11GB headroomest. 71.5 tok/s
Mac Studio, M5 Max, 48GB$3,099
Fits25GB headroomest. 24.5 tok/s
Mac Studio, M5 Max, 64GB$3,499
Fits39GB headroomest. 24.5 tok/s
Ascent GX10, GB10 Grace Blackwell, 128GB$3,999.99
Fits101GB headroomest. 10.9 tok/s
DGX Spark, GB10 Grace Blackwell, 128GB$4,699
Fits101GB headroomest. 10.9 tok/s
Mac Studio, M5 Max, 128GB$5,099
Fits100GB headroomest. 24.5 tok/s
Mac Studio, M5 Ultra, 96GB$5,499
Fits69GB headroomest. 47.8 tok/s
Mac Studio, M5 Ultra, 256GB$9,499
Fits228GB headroomest. 47.8 tok/s
Mac Studio, M5 Ultra, 512GB—
Fits484GB headroomest. 47.8 tok/s

Config	Memory	Price	$/GB	Fit	Headroom	Est. speed
Mac mini, M6	16GB	$1,099	$69	Won't fit	−7GB	—
Mac mini, M6	24GB	$1,299	$54	Tight	1GB	6.8 tok/s
Mac mini, M6	32GB	$1,499	$47	Fits	9GB	6.8 tok/s
Mac mini, M5 Pro	24GB	$1,699	$71	Tight	1GB	12.2 tok/s
Mac mini, M5 Pro	48GB	$2,499	$52	Fits	25GB	12.2 tok/s
Mac Studio, M5 Max	36GB	$2,499	$69	Fits	13GB	24.5 tok/s
Mac mini, M5 Pro	64GB	$2,899	$45	Fits	39GB	12.2 tok/s
GeForce RTX 5090	32GB	$2,999.99+host	$94	Fits	11GB	71.5 tok/s
Mac Studio, M5 Max	48GB	$3,099	$65	Fits	25GB	24.5 tok/s
Mac Studio, M5 Max	64GB	$3,499	$55	Fits	39GB	24.5 tok/s
Ascent GX10, GB10 Grace Blackwell	128GB	$3,999.99	$31	Fits	101GB	10.9 tok/s
DGX Spark, GB10 Grace Blackwell	128GB	$4,699	$37	Fits	101GB	10.9 tok/s
Mac Studio, M5 Max	128GB	$5,099	$40	Fits	100GB	24.5 tok/s
Mac Studio, M5 Ultra	96GB	$5,499	$57	Fits	69GB	47.8 tok/s
Mac Studio, M5 Ultra	256GB	$9,499	$37	Fits	228GB	47.8 tok/s
Mac Studio, M5 Ultra	512GB	—	—	Fits	484GB	47.8 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 = 27B × 0.604 B/param (4.83 bits) + 0.254 GB/1K × 8K K/V + 1.2 GB runtime
speed = bandwidth × 0.65 ÷ (27B active × 0.604 B/param) = 16.3 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

Qwen3-32BQ5_K_M, est. 4.9 tok/s
Gemma-3-27BQ6_K, est. 5.0 tok/s
gpt-oss-20bQ8_0, est. 22.2 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.