DGX Spark x3 vs DGX Spark x4

Best sourced output tok/s per model (single-stream generation) — engine and version, quantization, context and a source link on every row. Hover a value for the quant.

DGX Spark x32 sourced runs · best 152.9 tok/s

DGX Spark x411 verified runs · best 190.0 tok/s

Specs and evidence

	DGX Spark x3	DGX Spark x4
VRAM	128 GB	128 GB
Sourced runs on the board	2	11
Models measured	2	3
Best output tok/s	152.9 tok/s	190.0 tok/s
Highest measured peak VRAM	—	—
Street price (latest snapshot)	—	—

Sourced rows only. Peak VRAM is recorded by the person who ran the benchmark, so it is a floor for what the card can hold, not a specification. Price snapshots.

Head-to-head

On the 2 models measured on both cards, DGX Spark x3 leads on 0, DGX Spark x4 on 2. The widest gap is on DeepSeek-V4.1-Flash: DGX Spark x3 46.0 tok/s vs DGX Spark x4 82.0 tok/s (DGX Spark x4 78% faster).

The measured matrix

Model	DGX Spark x3	DGX Spark x4
DeepSeek-V4.1-Flash-1017	152.9 EXL3 3.5bpw	190.0 EXL3 3.5bpw
GLM-5.3-Flash	—	124.0 EXL3 4bpw
DeepSeek-V4.1-Flash	46.0 EXL3 3.5bpw	82.0 EXL3 3.5bpw

Head-to-head (models tested on both): DGX Spark x3 0 · DGX Spark x4 2 · tied 0 · data as of Oct 30, 2026. Full rules: methodology.

Related reading: RTX 5090 vs 3090 · quantization · Value index · Full board