Builds / 4x Nvidia GB10 (unified, DGX Spark)
4x DGX Spark (GB10) running Step 5 Preview with tensor parallel over RoCE
Source: github.com ↗📏 How measured: reported by the original source; RigBase re-checked the numbers at that source
- GPU(s)
- 4x Nvidia GB10 (unified, DGX Spark)
- VRAM / unified memory (GB)
- 512
- CPU
- not stated
- Motherboard
- not stated
- RAM
- 128 GB unified per node
- PSU
- not stated
- Other
- not stated
- Reported total cost
- not stated
- Power
- not stated
Measured speed (generation tokens/sec)
| Model | Quant | Engine | tok/s | Details |
|---|---|---|---|---|
| Step 5 Preview | EXL3 (experts 4.01 bpw, 341.4 GB total) | vLLM, TP=4 over RoCE, MTP (2 draft tokens) | 25 | ⚡ spec-decode (MTP)prefill 1,365 tok/scontext: 8K📏 ~25–30 tok/s single stream (published range); prefill 1,365–1,565 tok/s warm at 8K; 66.5 tok/s aggregate with 4 streams |
Measured 2026-10-04 to 06 by the repository author. tok/s shown is the low end of the published single-stream range.
Source: https://github.com/0xSero/Step-5-Preview-Four-Sparks
Blank fields = not stated in the source. We never estimate build data; GPU street prices above are separately sourced and dated.
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)
Quick answers: Can 4x DGX Spark (GB10) run Step 5 Preview?