IT
Prenoti una chiamata

Local AI · Recipe · 1× DGX Spark

Qwen3.8 27B MLX 4-bit with TensorFold on 1× DGX Spark

Qwen3.8 27B MLX 4-bit with TensorFold on 1× DGX Spark: 58.3 tok/s according to github.com (dataset as of Sep 29, 2026).

by MiaAI-Lab

58.3tok/sEveryday

1 request, prose prompt, with DFlash2 (per recipe)

Source: github.com
Intelligence (original model) · Artificial Analysis34with thinkingno thinking 20

Install: Qwen3.8 27B MLX 4-bit with TensorFold on 1× DGX Spark

Command
git clone https://github.com/MiaAI-Lab/Qwen3.8-27B-DGX-Spark-TensorFold.git && cd Qwen3.8-27B-DGX-Spark-TensorFold && ./start.sh

Full instructions in the README Repo

Engine

Engine
TensorFold v0.6.0 (c464617) + 5 Patches, Image ghcr.io/miaai-lab/qwen3.8-27b-dgx-spark-tensorfold:v0.6.0, DFlash2-Drafts
Quantization
MLX affine 4-bit (g64) + FP8 KV, FP8-Prefill
Model family
Qwen3.8
Context
262,144
Parameters
27B dense
Creator
MiaAI-Lab
GitHub stars
24
Repo updated
Oct 1, 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)58.3 tok/sProsa, sparkDash; TTFT 100 ms; eigener Fixed-Seed-Lauf: 48,6 Source 

Decode · parallel requests

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

  • 8 parallel requests323 tok/sPeak (decode) · Code ×8 aggregiert (53,3 pro Request); Code ×1 136,9; Prosa ×8 220,6 Source 

Prefill by context

  • 8K1,793 tok/s16k 1.785,3 / 32k 1.633,3 / 64k 1.415,9 / 128k 1.117,8 tok/s Source 

Time to first token

Shorter is better.

What you need

Hardware
1 × NVIDIA DGX Spark (GB10)
Weights
Engine
TensorFold v0.6.0 (c464617) + 5 Patches, Image ghcr.io/miaai-lab/qwen3.8-27b-dgx-spark-tensorfold:v0.6.0, DFlash2-Drafts
Context
262,144 tokens

Notes

What matters before you rebuild it.

  • Custom kernel required
  • vision

Sources

Related recipes

← 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