Local AI · Recipe · 2× DGX Spark

Qwen3.8-Flash-Next NVFP4 (hibrid48, kit v4.1) with vLLM on 2× DGX Spark

Qwen3.8-Flash-Next NVFP4 (hibrid48, kit v4.1) with vLLM on 2× DGX Spark: 121 tok/s according to github.com (dataset as of Nov 3, 2026).

by myllmbox

121tok/sPeak

1 request, code prompt

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

Engine

Engine
vLLM 0.30 (Kit v4.1, Image v6; `git checkout v4.1`) — main-Default ist seit v5.2 INT4-AutoRound
Quantization
hibrid48: NVFP4-Output-Head, bf16-KV (41G/Box-Pin), halbe NVFP4-n-gram-Tabelle pro Box, Marlin MoE
Model family
Qwen3.8-Flash-Next
Context
262,144
Creator
myllmbox
GitHub stars
132
Repo updated
Oct 29, 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

  • Peak (decode)121 tok/sSingle-stream-Peak, Code-Phase eines Thinking-on-Requests (auf dem v4-Image vor der Ladder gemessen); Average desselben Requests ~81; v5.2-README nennt für hibrid48 nur Schätzwerte (~120 tok/s bei 1 Stream, ~10 % langsamer als INT4) Source 

What you need

Hardware
2 × NVIDIA DGX Spark (GB10)
Engine
vLLM 0.30 (Kit v4.1, Image v6; `git checkout v4.1`) — main-Default ist seit v5.2 INT4-AutoRound
Context
262,144 tokens

Related recipes

2× DGX Spark
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
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.