Gemma 4 26B-A4B hardware requirements

Custom license · Gemma — check terms before commercial useChatVision

About the model

Parameters: 26.54B total · 4B active (MoE)
Attention: Sliding-window attention
Max context: 262,144 tokens
Released: 2026-04-01
Model card: Gemma 4 26B-A4B on Hugging Face

GGUF files

Tracked quantizations of Gemma 4 26B-A4B, smallest first
Quant	File size	Quality	Published by
UD-Q2_K_XL	10.55 GB	Heavy loss	Community GGUF · unsloth
Q4_K_Mbaseline	16.95 GB	Good balance	Community GGUF · unsloth
Q8_0	26.86 GB	Near-lossless	Community GGUF · unsloth

File sizes do not include the vision projector (mmproj) that image input needs.

Some repositories publish only Unsloth Dynamic (UD) variants; sizes are those of the file you would download.

Memory needed by quant and context

Weights + f16 KV cache + compute buffer, in GB. Add about 0.6 GB if the GPU also drives your display on Windows.
Quant	4k	8k	16k	32k	64k	128k	256k
UD-Q2_K_XL	11.3	11.5	11.9	12.7	14.3	17.6	24.2
Q4_K_M	17.7	17.9	18.3	19.1	20.7	24.0	30.6
Q8_0	27.6	27.8	28.2	29.0	30.6	33.9	40.5

Which GPUs can run Gemma 4 26B-A4B?

How to read the verdicts

Runs great: Fully on the GPU at 20 tok/s or more
Runs well: Fully on the GPU at 8–20 tok/s; MoE experts in system RAM at 20 tok/s or more; or at least 90% on the GPU at 8 tok/s or more
Runs slowly: 2–8 tok/s; CPU-only; less than 90% on the GPU; or MoE experts in system RAM below 20 tok/s
Won't run: Does not fit, or under 2 tok/s

Dense models split between GPU and CPU slow down sharply — the CPU side sets the pace. MoE models that keep only their experts in system RAM degrade far more gently.

Sort

Scroll sideways to see every column.

54 GPUs, Macs and CPU setups at Q4_K_M with 8k context, sorted by verdict and speed
Hardware	Verdict	Speed	Quant	Memory	Runs as	Context	Notes	Price
GeForce RTX 5090	Runs great	est. 235.0 tok/s188.0–282.0calibrated estimate ±20%	Q4_K_M	18.5 / 32.0 GB	Full GPU	up to 256k	—	$4,200 street (as of 2026-08-09)
Radeon RX 7900 XTX	Runs great	est. 166.1 tok/s132.9–199.3calibrated estimate ±20%	Q4_K_M	18.5 / 24.0 GB	Full GPU	up to 64k	—	$999 launch MSRP
GeForce RTX 3090 Ti	Runs great	est. 156.9 tok/s125.6–188.3calibrated estimate ±20%	Q4_K_M	18.5 / 24.0 GB	Full GPU	up to 64k	—	$1,999 launch MSRP
GeForce RTX 4090	Runs great	est. 156.9 tok/s125.6–188.3calibrated estimate ±20%	Q4_K_M	18.5 / 24.0 GB	Full GPU	up to 64k	—	$2,755 street (as of 2026-08-09)
GeForce RTX 3090	Runs great	est. 145.7 tok/s116.6–174.9calibrated estimate ±20%	Q4_K_M	18.5 / 24.0 GB	Full GPU	up to 64k	—	$1,050 street (as of 2026-08-09)
GeForce RTX 5090 Laptop	Runs great	est. 139.5 tok/s111.6–167.4calibrated estimate ±20%	Q4_K_M	18.5 / 24.0 GB	Full GPU	up to 64k	—	—
Radeon RX 7900 XT	Runs great	est. 138.4 tok/s110.7–166.1calibrated estimate ±20%	Q4_K_M	18.5 / 20.0 GB	Full GPU	up to 32k	—	$899 launch MSRP
Apple M3 Ultra · 96 GB	Runs great	est. 82.2 tok/s65.7–98.6calibrated estimate ±20%	Q4_K_M	17.9 / 67.2 GB	Unified memory	up to 128k	—	—
Apple M2 Ultra · 64 GB	Runs great	est. 80.3 tok/s64.2–96.3calibrated estimate ±20%	Q4_K_M	17.9 / 44.8 GB	Unified memory	up to 128k	—	—
Apple M5 Max (40-core GPU) · 48 GB	Runs great	est. 63.7 tok/s51.0–76.5calibrated estimate ±20%	Q4_K_M	17.9 / 33.6 GB	Unified memory	up to 128k	—	—
Apple M4 Max (40-core GPU) · 48 GB	Runs great	est. 56.7 tok/s45.3–68.0calibrated estimate ±20%	Q4_K_M	17.9 / 33.6 GB	Unified memory	up to 128k	—	—
Apple M5 Max (32-core GPU) · 36 GB	Runs great	est. 47.7 tok/s38.2–57.3calibrated estimate ±20%	Q4_K_M	17.9 / 25.2 GB	Unified memory	up to 64k	—	—
Ryzen AI Max+ 395 (Strix Halo) · 32 GB	Runs great	est. 46.9 tok/s37.6–56.3calibrated estimate ±20%	Q4_K_M	17.9 / 22.4 GB	Unified memory	up to 64k	—	$1,999 launch MSRP
Apple M4 Max (32-core GPU) · 36 GB	Runs great	est. 42.6 tok/s34.0–51.1calibrated estimate ±20%	Q4_K_M	17.9 / 25.2 GB	Unified memory	up to 64k	—	—
NVIDIA DGX Spark · 128 GB	Runs great	est. 37.8 tok/s30.2–45.3calibrated estimate ±20%	Q4_K_M	17.9 / 89.6 GB	Unified memory	up to 64k	—	$3,999 launch MSRP
Apple M5 Pro · 48 GB	Runs great	est. 31.9 tok/s25.5–38.2calibrated estimate ±20%	Q4_K_M	17.9 / 33.6 GB	Unified memory	up to 32k	—	—
Apple M4 Pro · 48 GB	Runs great	est. 28.3 tok/s22.7–34.0calibrated estimate ±20%	Q4_K_M	17.9 / 33.6 GB	Unified memory	up to 32k	—	—
Apple M5 · 32 GB	Runs well	est. 15.9 tok/s12.8–19.1calibrated estimate ±20%	Q4_K_M	17.9 / 22.4 GB	Unified memory	up to 64k	UD-Q2_K_XL (heavy quality loss): Runs great	—
Apple M4 · 32 GB	Runs well	est. 12.5 tok/s10.0–14.9calibrated estimate ±20%	Q4_K_M	17.9 / 22.4 GB	Unified memory	up to 32k	—	—
GeForce RTX 5080	Runs slowly	est. 60.2 tok/s42.1–78.3theoretical estimate ±30%	Q4_K_M	16.0 / 16.0 GB + 2.5 GB RAM	Partial offload	up to 256k	Less than 90% on the GPU — a GPU/CPU split is capped at Runs slowly UD-Q2_K_XL (heavy quality loss): Runs great	$1,256 street (as of 2026-08-09)
GeForce RTX 5070 Ti	Runs slowly	est. 58.8 tok/s41.1–76.4theoretical estimate ±30%	Q4_K_M	16.0 / 16.0 GB + 2.5 GB RAM	Partial offload	up to 256k	Less than 90% on the GPU — a GPU/CPU split is capped at Runs slowly UD-Q2_K_XL (heavy quality loss): Runs great	$949 street (as of 2026-08-09)
GeForce RTX 5080 Laptop	Runs slowly	est. 58.8 tok/s41.1–76.4theoretical estimate ±30%	Q4_K_M	16.0 / 16.0 GB + 2.5 GB RAM	Partial offload	up to 256k	Less than 90% on the GPU — a GPU/CPU split is capped at Runs slowly UD-Q2_K_XL (heavy quality loss): Runs great	—
GeForce RTX 4080 Super	Runs slowly	est. 54.5 tok/s38.1–70.8theoretical estimate ±30%	Q4_K_M	16.0 / 16.0 GB + 2.5 GB RAM	Partial offload	up to 256k	Less than 90% on the GPU — a GPU/CPU split is capped at Runs slowly UD-Q2_K_XL (heavy quality loss): Runs great	$999 launch MSRP
GeForce RTX 4080	Runs slowly	est. 53.9 tok/s37.7–70.1theoretical estimate ±30%	Q4_K_M	16.0 / 16.0 GB + 2.5 GB RAM	Partial offload	up to 256k	Less than 90% on the GPU — a GPU/CPU split is capped at Runs slowly UD-Q2_K_XL (heavy quality loss): Runs great	$1,199 launch MSRP
Radeon RX 9070	Runs slowly	est. 53.9 tok/s37.7–70.1theoretical estimate ±30%	Q4_K_M	16.0 / 16.0 GB + 2.5 GB RAM	Partial offload	up to 256k	Less than 90% on the GPU — a GPU/CPU split is capped at Runs slowly UD-Q2_K_XL (heavy quality loss): Runs great	$639 street (as of 2026-08-09)
Radeon RX 9070 XT	Runs slowly	est. 53.9 tok/s37.7–70.1theoretical estimate ±30%	Q4_K_M	16.0 / 16.0 GB + 2.5 GB RAM	Partial offload	up to 256k	Less than 90% on the GPU — a GPU/CPU split is capped at Runs slowly UD-Q2_K_XL (heavy quality loss): Runs great	$689 street (as of 2026-08-09)
Radeon RX 7800 XT	Runs slowly	est. 53.2 tok/s37.2–69.1theoretical estimate ±30%	Q4_K_M	16.0 / 16.0 GB + 2.5 GB RAM	Partial offload	up to 256k	Less than 90% on the GPU — a GPU/CPU split is capped at Runs slowly UD-Q2_K_XL (heavy quality loss): Runs great	$499 launch MSRP
GeForce RTX 4070 Ti Super	Runs slowly	est. 52.5 tok/s36.7–68.2theoretical estimate ±30%	Q4_K_M	16.0 / 16.0 GB + 2.5 GB RAM	Partial offload	up to 256k	Less than 90% on the GPU — a GPU/CPU split is capped at Runs slowly UD-Q2_K_XL (heavy quality loss): Runs great	$799 launch MSRP
GeForce RTX 4090 Laptop	Runs slowly	est. 49.0 tok/s34.3–63.7theoretical estimate ±30%	Q4_K_M	16.0 / 16.0 GB + 2.5 GB RAM	Partial offload	up to 256k	Less than 90% on the GPU — a GPU/CPU split is capped at Runs slowly UD-Q2_K_XL (heavy quality loss): Runs great	—
GeForce RTX 5060 Ti 16GB	Runs slowly	est. 43.2 tok/s30.2–56.2theoretical estimate ±30%	Q4_K_M	16.0 / 16.0 GB + 2.5 GB RAM	Partial offload	up to 256k	Less than 90% on the GPU — a GPU/CPU split is capped at Runs slowly UD-Q2_K_XL (heavy quality loss): Runs great	$569 street (as of 2026-08-09)
Radeon RX 9060 XT 16GB	Runs slowly	est. 38.1 tok/s26.7–49.6theoretical estimate ±30%	Q4_K_M	16.0 / 16.0 GB + 2.5 GB RAM	Partial offload	up to 256k	Less than 90% on the GPU — a GPU/CPU split is capped at Runs slowly UD-Q2_K_XL (heavy quality loss): Runs great	$449 street (as of 2026-08-09)
GeForce RTX 4060 Ti 16GB	Runs slowly	est. 33.4 tok/s23.4–43.4theoretical estimate ±30%	Q4_K_M	16.0 / 16.0 GB + 2.5 GB RAM	Partial offload	up to 256k	Less than 90% on the GPU — a GPU/CPU split is capped at Runs slowly UD-Q2_K_XL (heavy quality loss): Runs great	$499 launch MSRP
GeForce RTX 3080 12GB	Runs slowly	est. 30.3 tok/s21.2–39.5theoretical estimate ±30%	Q4_K_M	12.0 / 12.0 GB + 6.5 GB RAM	Partial offload	up to 128k	Less than 90% on the GPU — a GPU/CPU split is capped at Runs slowly UD-Q2_K_XL (heavy quality loss): Runs well	$799 launch MSRP
GeForce RTX 5070	Runs slowly	est. 29.0 tok/s20.3–37.7theoretical estimate ±30%	Q4_K_M	12.0 / 12.0 GB + 6.5 GB RAM	Partial offload	up to 128k	Less than 90% on the GPU — a GPU/CPU split is capped at Runs slowly UD-Q2_K_XL (heavy quality loss): Runs well	$629 street (as of 2026-08-09)
GeForce RTX 5070 Ti Laptop	Runs slowly	est. 29.0 tok/s20.3–37.7theoretical estimate ±30%	Q4_K_M	12.0 / 12.0 GB + 6.5 GB RAM	Partial offload	up to 128k	Less than 90% on the GPU — a GPU/CPU split is capped at Runs slowly UD-Q2_K_XL (heavy quality loss): Runs well	—
Intel Arc B580	Runs slowly	est. 28.4 tok/s19.9–36.9theoretical estimate ±30%	Q4_K_M	12.0 / 12.0 GB + 6.5 GB RAM	Partial offload	up to 128k	Less than 90% on the GPU — a GPU/CPU split is capped at Runs slowly UD-Q2_K_XL (heavy quality loss): Runs well	$290 street (as of 2026-08-09)
GeForce RTX 4070	Runs slowly	est. 27.4 tok/s19.2–35.6theoretical estimate ±30%	Q4_K_M	12.0 / 12.0 GB + 6.5 GB RAM	Partial offload	up to 128k	Less than 90% on the GPU — a GPU/CPU split is capped at Runs slowly UD-Q2_K_XL (heavy quality loss): Runs well	$599 launch MSRP
GeForce RTX 4070 Super	Runs slowly	est. 27.4 tok/s19.2–35.6theoretical estimate ±30%	Q4_K_M	12.0 / 12.0 GB + 6.5 GB RAM	Partial offload	up to 128k	Less than 90% on the GPU — a GPU/CPU split is capped at Runs slowly UD-Q2_K_XL (heavy quality loss): Runs well	$599 launch MSRP
GeForce RTX 4080 Laptop	Runs slowly	est. 26.5 tok/s18.5–34.4theoretical estimate ±30%	Q4_K_M	12.0 / 12.0 GB + 6.5 GB RAM	Partial offload	up to 128k	Less than 90% on the GPU — a GPU/CPU split is capped at Runs slowly UD-Q2_K_XL (heavy quality loss): Runs well	—
GeForce RTX 2080 Ti	Runs slowly	est. 25.7 tok/s18.0–33.5theoretical estimate ±30%	Q4_K_M	11.0 / 11.0 GB + 7.5 GB RAM	Partial offload	up to 128k	Less than 90% on the GPU — a GPU/CPU split is capped at Runs slowly	$999 launch MSRP
GeForce RTX 3060 12GB	Runs slowly	est. 25.2 tok/s17.7–32.8theoretical estimate ±30%	Q4_K_M	12.0 / 12.0 GB + 6.5 GB RAM	Partial offload	up to 128k	Less than 90% on the GPU — a GPU/CPU split is capped at Runs slowly UD-Q2_K_XL (heavy quality loss): Runs well	$250 street (as of 2026-08-09)
GeForce GTX 1080 Ti	Runs slowly	est. 24.7 tok/s17.3–32.1theoretical estimate ±30%	Q4_K_M	11.0 / 11.0 GB + 7.5 GB RAM	Partial offload	up to 128k	Less than 90% on the GPU — a GPU/CPU split is capped at Runs slowly	$699 launch MSRP
GeForce RTX 3080 10GB	Runs slowly	est. 24.0 tok/s16.8–31.2theoretical estimate ±30%	Q4_K_M	10.0 / 10.0 GB + 8.5 GB RAM	Partial offload	up to 128k	Less than 90% on the GPU — a GPU/CPU split is capped at Runs slowly	$699 launch MSRP
GeForce RTX 3070	Runs slowly	est. 19.3 tok/s13.5–25.1theoretical estimate ±30%	Q4_K_M	8.0 / 8.0 GB + 10.5 GB RAM	Partial offload	up to 128k	Less than 90% on the GPU — a GPU/CPU split is capped at Runs slowly	$499 launch MSRP
GeForce RTX 5060	Runs slowly	est. 19.3 tok/s13.5–25.1theoretical estimate ±30%	Q4_K_M	8.0 / 8.0 GB + 10.5 GB RAM	Partial offload	up to 128k	Less than 90% on the GPU — a GPU/CPU split is capped at Runs slowly	$339 street (as of 2026-08-09)
GeForce RTX 5060 Ti 8GB	Runs slowly	est. 19.3 tok/s13.5–25.1theoretical estimate ±30%	Q4_K_M	8.0 / 8.0 GB + 10.5 GB RAM	Partial offload	up to 128k	Less than 90% on the GPU — a GPU/CPU split is capped at Runs slowly	$429 street (as of 2026-08-09)
GeForce RTX 4060 Ti 8GB	Runs slowly	est. 18.2 tok/s12.8–23.7theoretical estimate ±30%	Q4_K_M	8.0 / 8.0 GB + 10.5 GB RAM	Partial offload	up to 128k	Less than 90% on the GPU — a GPU/CPU split is capped at Runs slowly	$399 launch MSRP
GeForce RTX 4060	Runs slowly	est. 18.1 tok/s12.7–23.5theoretical estimate ±30%	Q4_K_M	8.0 / 8.0 GB + 10.5 GB RAM	Partial offload	up to 128k	Less than 90% on the GPU — a GPU/CPU split is capped at Runs slowly	$299 launch MSRP
GeForce RTX 4060 Laptop	Runs slowly	est. 17.9 tok/s12.5–23.3theoretical estimate ±30%	Q4_K_M	8.0 / 8.0 GB + 10.5 GB RAM	Partial offload	up to 128k	Less than 90% on the GPU — a GPU/CPU split is capped at Runs slowly	—
GeForce RTX 4070 Laptop	Runs slowly	est. 17.9 tok/s12.5–23.3theoretical estimate ±30%	Q4_K_M	8.0 / 8.0 GB + 10.5 GB RAM	Partial offload	up to 128k	Less than 90% on the GPU — a GPU/CPU split is capped at Runs slowly	—
GeForce RTX 3050 8GB	Runs slowly	est. 17.5 tok/s12.2–22.7theoretical estimate ±30%	Q4_K_M	8.0 / 8.0 GB + 10.5 GB RAM	Partial offload	up to 128k	Less than 90% on the GPU — a GPU/CPU split is capped at Runs slowly	$249 launch MSRP
DDR5-6000 dual-channel CPU · 32 GB	Runs slowly	est. 14.3 tok/s11.4–17.1calibrated estimate ±20%	Q4_K_M	17.3 GB RAM	CPU only	up to 256k	CPU inference — capped at Runs slowly	—
DDR5-5600 dual-channel CPU · 32 GB	Runs slowly	est. 13.3 tok/s10.7–16.0calibrated estimate ±20%	Q4_K_M	17.3 GB RAM	CPU only	up to 256k	CPU inference — capped at Runs slowly	—
DDR4-3200 dual-channel CPU · 32 GB	Runs slowly	est. 7.6 tok/s6.1–9.1calibrated estimate ±20%	Q4_K_M	17.3 GB RAM	CPU only	up to 128k	—	—

Cheapest GPUs that run it

Runs greatGeForce RTX 3090$1,050 street (as of 2026-08-09), est. 145.7 tok/s (116.6–174.9, calibrated estimate ±20%).

Measured results for Gemma 4 26B-A4B

No public measurements for this model yet.

Frequently asked questions

How much VRAM does Gemma 4 26B-A4B need?

At Q4_K_M the weights are 16.95 GB; with 8k context the total is about 17.9 GB (KV cache 0.3 GB, compute buffer 0.6 GB). The smallest tracked file (UD-Q2_K_XL (heavy quality loss)) needs about 11.5 GB.

What is the cheapest GPU that runs Gemma 4 26B-A4B well?

Cheapest that runs it great: GeForce RTX 3090 at $1,050 (street price as of 2026-08-09), est. 145.7 tok/s (116.6–174.9, calibrated estimate ±20%).

Can Gemma 4 26B-A4B run on an 8 GB, 16 GB or 24 GB GPU?

At Q4_K_M with 8k context: GeForce RTX 4060 — Runs slowly, est. 18.1 tok/s (12.7–23.5, theoretical estimate ±30%); GeForce RTX 5060 Ti 16GB — Runs slowly, est. 43.2 tok/s (30.2–56.2, theoretical estimate ±30%); GeForce RTX 3090 — Runs great, est. 145.7 tok/s (116.6–174.9, calibrated estimate ±20%).

Speeds are estimates from memory bandwidth, calibrated against public benchmarks, and each one comes with an error band and a confidence label. Real results vary with drivers, backend, context length and thermals.