Builds / 2x Nvidia GB10 (unified, DGX Spark)

2x DGX Spark: MiniMax M3 W4A16 GPTQ on vLLM TP2, 31.8 tok/s code (Tech2wild)

Last updated: 28 Oct 2026

Source: github.com ↗More source: github.com ↗📏 How measured: reported by the original source; RigBase re-checked the numbers at that source

Legacy models only. Every model measured on this rig now has a newer successor: MiniMax M3 → MiniMax M3.1 Flash. The hardware numbers are still a useful reference.

GPU(s): 2x Nvidia GB10 (unified, DGX Spark)
VRAM / unified memory (GB): 256
CPU: Nvidia GB10 Grace (Arm) per node
Amazon · eBay (used) · Newegg
Motherboard: not stated
RAM: 128 GB LPDDR5X unified per Spark (256 GB total)
PSU: not stated
Other: 2 NVIDIA DGX Sparks, multi-node
Reported total cost: not stated
Power: not stated

Measured speed (generation tokens/sec)

Model	Quant	Engine	tok/s	Prompt processing (tok/s)	Details
MiniMax M3 legacy	W4A16 GPTQ + 4-bit KV	vLLM TP2, 2 nodes, 1 stream	31.8	not stated	context: 196K📏 Single-stream code decode as published in Tech2wild's recipe README.

Numbers by Tech2wild (tonyd2wild). RigBase read the GitHub README on 2026-10-28 and checked each figure against the text. RigBase did not re-run it. Speculative decoding (MTP/DFlash/DSpark) inflates predictable text; these rows prefer prose/code over counting peaks. Recipe license: Apache-2.0.

Source: https://github.com/tonyd2wild/MiniMax-M3-2x-DGX-Spark-36-tok-s
https://github.com/tonyd2wild/MiniMax-M3-2x-DGX-Spark-36-tok-s/blob/main/README.md

Blank fields = not stated in the source. We never estimate build data; GPU street prices above are separately sourced and dated.

Embed this result

Copy into a README or blog. Links back to this build. See badge docs (CC BY 4.0 attribution). The SVG does not invent speeds — numbers stay on the build page.

Markdown

[![Measured on RigBase](https://getrigbase.com/badge/tony-2x-dgx-spark-minimax-m3-2x-dgx-spark-36-tok-s.svg)](https://getrigbase.com/builds/tony-2x-dgx-spark-minimax-m3-2x-dgx-spark-36-tok-s/)

HTML

<a href="https://getrigbase.com/builds/tony-2x-dgx-spark-minimax-m3-2x-dgx-spark-36-tok-s/" rel="noopener"><img src="https://getrigbase.com/badge/tony-2x-dgx-spark-minimax-m3-2x-dgx-spark-36-tok-s.svg" alt="Measured on RigBase" height="20"/></a>

Preview: · How to use

Rent before you buy: try this model on a cloud GPU for a few dollars first: Runpod (referral link; sign up with Google to get the new-user bonus)