EN
Book a call

Local AI · Recipe · 2× DGX Spark

Qwen3.8-Flash-Next NVFP4/MXFP8 CSF with SparkRing on 2× DGX Spark

Qwen3.8-Flash-Next NVFP4/MXFP8 CSF with SparkRing on 2× DGX Spark: 68 tok/s according to github.com (dataset as of Nov 3, 2026).

by fujitsupolycom

68tok/sEveryday

1 request, generation at 32K context (speed table, pair), thinking on, xhigh, with MTP3 (per recipe)

Source: github.com
Intelligence (original model) · Artificial Analysis40with thinking

Install: Qwen3.8-Flash-Next NVFP4/MXFP8 CSF with SparkRing on 2× DGX Spark

Command
curl -fsSL https://fujitsupolycom.github.io/sparkring/install.sh | bash -s -- --profile qwen38-flash-next-tp2

Full instructions in the README Repo

Engine

Engine
vLLM (local-inference-lab Fork) via SparkRing (SIRCL-Collectives, switchless RDMA ueber ConnectX-7), Image Release 2026.10.2
Quantization
NVFP4/MXFP8 CSF QAD (Local Inference Lab)
Model family
Qwen3.8-Flash-Next
Context
262,144
Parameters
48 Layer, 512 Experten (MoE, 10 aktiv); 102,2 GB Safetensors
Creator
fujitsupolycom
GitHub stars
102
Repo updated
Nov 2, 2026

Measurements

Every sourced value of this recipe, each with its condition and source. Bars relative to the largest value in the group.

Decode · 1 request

  • Everyday (decode)68 tok/scontext 32K · 1 Stream, 32K Kontext, Paar (Ring B, zweite Messung desselben Starts, Ring sonst idle), Release-Image 2026.10.2, GPU-Takt fest 2.418 MHz; Thinking on, xhigh; MTP3 Source 

Decode · parallel requests

Sums over all streams are marked; per request it is correspondingly less.

  • sum over 8 streams219 tok/sPeak (decode) · context 32K · Aggregate of 8 streams at 32K context, pair; thinking on, xhigh Source 

Prefill by context

  • 32K3,716 tok/sEin Prompt mit ca. 32.200 Tokens (Client-Messung bis zum ersten Token) Source 

What you need

Hardware
2 × NVIDIA DGX Spark (GB10)
Engine
vLLM (local-inference-lab Fork) via SparkRing (SIRCL-Collectives, switchless RDMA ueber ConnectX-7), Image Release 2026.10.2
Context
262,144 tokens

Notes

What matters before you rebuild it.

  • Custom kernel required
  • License unclear

Sources

Related recipes

2× DGX Spark

Qwen3.8-Flash-Next 125B INT4-AutoRound with TensorFold on 2× DGX Spark (1M context)

by MiaAI-Lab

89.3tok/sEveryday

1 request, prose prompt, with MTP (per recipe) Source 

Intelligenceno independent value
  • Custom kernel required
  • License unclear

INT4-AutoRound (GPTQ int4, Top-5-Experten) + FP8 n-gram-Tabelle, FP8 KVTensorFold tensorfold-native (Zig, zig-flashnext db281878 + patches), Image ghcr.io/miaai-lab/qwen3.8-flash-dual-dgx-sparks-tensorfold:zig-db28187-f3e1300f4f62, TP=2 über CX7/RoCE, MTP + Copy-Drafts (QUANT=int4ar)Weights Repo Updated Oct 26, 2026

Details
2× DGX Spark

Qwen3.8-Flash-Next INT4-AutoRound (v5.2) with vLLM on 2× DGX Spark

by myllmbox

74tok/sEveryday

1 request, prose prompt (long story), thinking off, with MTP Source 

Intelligence40with thinking
  • Custom kernel required
  • License unclear

INT4-AutoRound (azampatti): int4-Experten, fp8-Seitenschichten; n-gram-Tabelle beim Laden fp8 → NVFP4, halb pro Box auf der GPU; KV-Pin 29G/Box (1.829.182 Tokens)vLLM 0.30 (Image myllmbox/qwen38-flash-next-cluster-vllm:v5.2; RecoverSSM, dynamischer Draft-Depth 3–7 in Full-CUDA-Graphs, sampled Drafts + Block-Verifikation, One-Shot-RoCE-All-Reduce, Dual-Rail, TP=2, 64 Seats) — Kit v5.2Weights Repo Updated Oct 29, 2026

Details

← Back to overview

Local AI in your company?

In a workshop we work out which models and which hardware fit your tasks, and build the first agent on your infrastructure.

See the AI workshopBack to overview
Cookies

With your consent we use Google Analytics (Google, USA) and our own audience measurement to understand how our website is used. Strictly necessary storage, such as your cookie choice, happens without consent. You can change your choice at any time via “Cookie settings” in the footer. Privacy Policy · Cookie Policy