How much memory does Gemma-4-26B-A4B at Q4_K_M need?

17GB. That is 15GB of weights at 4.83 bits per weight, 0.3GB of K/V cache at 8K context, and 1.2GB of runtime overhead.

Which machine is cheapest for Gemma-4-26B-A4B at Q4_K_M?

Mac mini, M6, 24GB, $1,299 at Best Buy. It fits with 4GB of headroom and decodes at an est. 37.0 tok/s.

Cheapest box to run Gemma-4-26B-A4B at Q4_K_M

Google’s Gemma-4-26B-A4B is 25.2B parameters, 3.8B of them read per token. 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 33 days ago

Quant

Context

Gemma-4-26B-A4B at Q4_K_M needs 17GB.

Cheapest Config that fits: Mac mini, M6, 24GB. Fits, 4GB headroom. Est. 37.0 tok/s.

Best offer today: $1,299 at Best Buy — in stock, checked 33 days ago.

Price history: 34 days recorded.

The 26B in the name is marketing; Google's own table reads 25.2B total and 3.8B active, and that is what we publish. Only the 5 global-attention layers cache unboundedly; the rest slide a 1024-token window.

Offers — Mac mini, M6, 24GB, 512GB

Best BuyApple's own price, to the cent.in stockchecked 33 days ago$1,299
ExpercomApple's own price, to the cent.in stockchecked 6 days ago$1,299
Applelist pricein stockchecked 32 minutes ago$1,299

Mac mini, M6, 16GB$1,099
Won't fit−4GB headroom
Mac mini, M6, 24GB$1,299
Fits4GB headroomest. 37.0 tok/s
Mac mini, M6, 32GB$1,499
Fits12GB headroomest. 37.0 tok/s
Mac mini, M5 Pro, 24GB$1,699
Fits4GB headroomest. 66.9 tok/s
Mac mini, M5 Pro, 48GB$2,499
Fits27GB headroomest. 66.9 tok/s
Mac Studio, M5 Max, 36GB$2,499
Fits16GB headroomest. 100 tok/s
Mac mini, M5 Pro, 64GB$2,899
Fits42GB headroomest. 66.9 tok/s
GeForce RTX 5090, 32GB$2,999.99+host
Fits14GB headroomest. 391 tok/s
Mac Studio, M4 Max, 64GB, refurbished$3,049
Fits42GB headroomest. 119 tok/s
Mac Studio, M5 Max, 48GB$3,099
Fits27GB headroomest. 134 tok/s
Mac Studio, M5 Max, 64GB$3,499
Fits42GB headroomest. 134 tok/s
Ascent GX10, GB10 Grace Blackwell, 128GB$3,999.99
Fits104GB headroomest. 59.5 tok/s
Mac Studio, M3 Ultra, 96GB, refurbished$4,329
Fits72GB headroomest. 178 tok/s
Mac Studio, M4 Max, 128GB, refurbished$4,409
Fits103GB headroomest. 119 tok/s
DGX Spark, GB10 Grace Blackwell, 128GB$4,699
Fits104GB headroomest. 59.5 tok/s
Mac Studio, M5 Max, 128GB$5,099
Fits103GB headroomest. 134 tok/s
Mac Studio, M5 Ultra, 96GB$5,499
Fits72GB headroomest. 262 tok/s
Mac Studio, M3 Ultra, 256GB, refurbished$7,729
Fits231GB headroomest. 178 tok/s
Mac Studio, M5 Ultra, 256GB$9,499
Fits231GB headroomest. 262 tok/s
Mac Studio, M5 Ultra, 512GB—
Fits487GB headroomest. 262 tok/s

Config	Memory	Price	$/GB	Fit	Headroom	Est. speed
Mac mini, M6	16GB	$1,099	$69	Won't fit	−4GB	—
Mac mini, M6	24GB	$1,299	$54	Fits	4GB	37.0 tok/s
Mac mini, M6	32GB	$1,499	$47	Fits	12GB	37.0 tok/s
Mac mini, M5 Pro	24GB	$1,699	$71	Fits	4GB	66.9 tok/s
Mac mini, M5 Pro	48GB	$2,499	$52	Fits	27GB	66.9 tok/s
Mac Studio, M5 Max	36GB	$2,499	$69	Fits	16GB	100 tok/s
Mac mini, M5 Pro	64GB	$2,899	$45	Fits	42GB	66.9 tok/s
GeForce RTX 5090	32GB	$2,999.99+host	$94	Fits	14GB	391 tok/s
Mac Studio, M4 Max, refurb	64GB	$3,049	$48	Fits	42GB	119 tok/s
Mac Studio, M5 Max	48GB	$3,099	$65	Fits	27GB	134 tok/s
Mac Studio, M5 Max	64GB	$3,499	$55	Fits	42GB	134 tok/s
Ascent GX10, GB10 Grace Blackwell	128GB	$3,999.99	$31	Fits	104GB	59.5 tok/s
Mac Studio, M3 Ultra, refurb	96GB	$4,329	$45	Fits	72GB	178 tok/s
Mac Studio, M4 Max, refurb	128GB	$4,409	$34	Fits	103GB	119 tok/s
DGX Spark, GB10 Grace Blackwell	128GB	$4,699	$37	Fits	104GB	59.5 tok/s
Mac Studio, M5 Max	128GB	$5,099	$40	Fits	103GB	134 tok/s
Mac Studio, M5 Ultra	96GB	$5,499	$57	Fits	72GB	262 tok/s
Mac Studio, M3 Ultra, refurb	256GB	$7,729	$30	Fits	231GB	178 tok/s
Mac Studio, M5 Ultra	256GB	$9,499	$37	Fits	231GB	262 tok/s
Mac Studio, M5 Ultra	512GB	—	—	Fits	487GB	262 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 = 25.2B × 0.604 B/param (4.83 bits) + 0.042 GB/1K × 8K K/V + 1.2 GB runtime
speed = bandwidth × 0.5 ÷ (3.8B active × 0.604 B/param) = 2.3 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 24GB runs

Qwen3.5-35B-A3BQ3_K_M, est. 58.0 tok/s
Qwen3-32BQ3_K_M, est. 7.1 tok/s
Gemma-4-31BQ3_K_M, est. 7.4 tok/s
Qwen3-30B-A3BQ4_K_M, est. 42.7 tok/s
Gemma-3-27BQ4_K_M, est. 6.8 tok/s
Qwen3.6-27BQ5_K_M, est. 5.8 tok/s

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

Bandwidth 170GB/s, usable memory 21GB of 24GB.