TokenAssemble

Can an NVIDIA DGX Spark run gemma-4-26B-A4B-it?

Yes. gemma-4-26B-A4B-it needs 29.3GB at Q8_0 with llama.cpp and 8K context; the NVIDIA DGX Spark has 128GB of unified memory (~90% GPU-addressable) (114.7GB usable) — grade A+, with decode speed in the interactive tier.

NVIDIA DGX Spark · gemma-4-26B-A4B-it · Q8_0

Fits — grade A+

InteractivebetaHybrid attention — KV computed as full-cache upper bound · MoE speed estimate — wide error bars · New hardware class — tier not yet calibrated
weights 26.9 GBKV cache 2.01 GBoverhead 0.44 GBpool 114.7 GBneeds 29.3 GB

assumes llama.cpp · batch 1 · 8K context · fp16 KV cache

$4,699 MSRP · as of 2026-07-11

Affiliate link — we may earn a commission. Verdicts are computed before any link is attached.

Every quant, graded

NVIDIA DGX Spark × gemma-4-26B-A4B-it, llama.cpp, 8K context.

QuantFile sizeFitSpeed tier
Q8_026.9 GBFits · A+interactive · beta
UD-Q5_K_M21.1 GBFits · A+interactive · beta
UD-Q4_K_M16.9 GBFits · A+interactive · beta
MXFP4_MOE16.6 GBFits · A+interactive · beta

gemma-4-26B-A4B-it on similar hardware

More models on the NVIDIA DGX Spark

Verdict computed from published specs and measured GGUF file sizes — see how the math works.