---
type: "Article"
title: "Qwen3.8-Flash-Next NVFP4 (NVIDIA QAD, FP4-Local) mit TensorFold auf 1× DGX Spark"
description: "Qwen3.8-Flash-Next NVFP4 (NVIDIA QAD, FP4-Local) mit TensorFold auf 1× DGX Spark: Rezept von MiaAI-Lab mit Tempo, Bedingung und Quelle je Zahl, Gewichten und Lizenz."
resource: "https://www.contextstudios.ai/de/lokale-ki/miaai-lab--qwen3.8-flash-next-fp4-qad-single-dgx-spark-tensorfold"
language: "de"
generated:
  by: "process:contextstudios-md/1"
  at: "2026-11-11T03:30:13.641Z"
status: "stable"
---

# Qwen3.8-Flash-Next NVFP4 (NVIDIA QAD, FP4-Local) mit TensorFold auf 1× DGX Spark

Qwen3.8-Flash-Next NVFP4 (NVIDIA QAD, FP4-Local) mit TensorFold auf 1× DGX Spark: 55,6 tok/s laut github.com (Datensatz-Stand: 3. Nov. 2026).

- Kreator: [MiaAI-Lab](https://www.contextstudios.ai/de/lokale-ki/kreatoren/miaai-lab.md)
- Engine: TensorFold tensorfold-native (Zig) v1.1.4 (eab85828) + Patches 0001–0017, Image tensorfold-qwen38:v1.1.4-eab8582 (GHCR v1.1.4-eab8582-4baf16137ecd) auf nvcr.io/nvidia/pytorch:26.07-py3, MTP-Drafts
- Quantisierung: NVIDIA FP4-QAD-Export (FP4-Local @b9ce18a6): Routed Experts NVFP4 (Gruppe 16), Attention/Shared Expert MXFP8, alle 512 Experten mit 10 pro Token; n-gram-Tabelle per pread von der SSD; FP8 KV
- Modellfamilie: Qwen3.8-Flash-Next
- Hardware: dgx-spark x1
- Kontext: 262144
- Artificial Analysis Intelligence Index: 39.8 ([source](https://artificialanalysis.ai/models/qwen3-8-flash-next))

## Alle Zahlen

- 55,6 tok/s (Alltag; Alltag, 1 Anfrage, Prosa-Prompt, mit MTP (laut Rezept)) — [source](https://github.com/MiaAI-Lab/Qwen3.8-Flash-Next-Single-DGX-Spark-TensorFold#performance)
- decode_tps: 55.6 (1 Anfrage) — [source](https://github.com/MiaAI-Lab/Qwen3.8-Flash-Next-Single-DGX-Spark-TensorFold#performance)
- decode_tps: 156.1 (5 parallele Anfragen) — [source](https://github.com/MiaAI-Lab/Qwen3.8-Flash-Next-Single-DGX-Spark-TensorFold#performance)
- prefill_tps: 2839 (1 Anfrage) — [source](https://github.com/MiaAI-Lab/Qwen3.8-Flash-Next-Single-DGX-Spark-TensorFold#performance)
- ttft_ms: 136 (1 Anfrage) — [source](https://github.com/MiaAI-Lab/Qwen3.8-Flash-Next-Single-DGX-Spark-TensorFold#performance)

## Gewichte

- [nvidia/Qwen3.8-Flash-Next-FP4-Local](https://huggingface.co/nvidia/Qwen3.8-Flash-Next-FP4-Local) (main, other)
- [Qwen/Qwen3.8-Flash-Next](https://huggingface.co/Qwen/Qwen3.8-Flash-Next) (base, other)

[https://github.com/MiaAI-Lab/Qwen3.8-Flash-Next-Single-DGX-Spark-TensorFold](https://github.com/MiaAI-Lab/Qwen3.8-Flash-Next-Single-DGX-Spark-TensorFold)

## Related

- [Lokale KI auf Ihrer Hardware.](https://www.contextstudios.ai/de/lokale-ki.md)