---
type: "Article"
title: "Qwen3.8-Flash-Next NVFP4 with vLLM on 2× DGX Spark (Weschera)"
description: "Qwen3.8-Flash-Next NVFP4 with vLLM on 2× DGX Spark (Weschera): recipe by Weschera with speed, condition and source for every number, weights and license."
resource: "https://www.contextstudios.ai/local-ai/weschera--qwen38-flash-next-nvfp4-nvidia-mtp3-2x-dgx-spark"
language: "en"
generated:
  by: "process:contextstudios-md/1"
  at: "2026-10-02T21:56:43.937Z"
status: "stable"
---

# Qwen3.8-Flash-Next NVFP4 with vLLM on 2× DGX Spark (Weschera)

Qwen3.8-Flash-Next NVFP4 with vLLM on 2× DGX Spark (Weschera): 33 tok/s according to github.com (dataset as of Sep 29, 2026).

- Creator: [Weschera](https://www.contextstudios.ai/local-ai/creators/weschera.md)
- Engine: vLLM
- Quantization: NVFP4 (ModelOpt MIXED_PRECISION: NVFP4 Experts, FP8 dense/PLE/MTP); BF16 KV (Pflicht — QSA lehnt fp8-KV ab)
- Model family: Qwen3.8-Flash-Next
- Hardware: dgx-spark x2
- Context: 131072

## All numbers

- 33 tok/s (Everyday; Everyday, 1 request, prose prompt, with MTP) — [source](https://github.com/Weschera/Qwen3.8-Flash-Next-NVFP4-2x-DGX-Spark/blob/main/README.md)
- decode_tps: 33 (1 request) — [source](https://github.com/Weschera/Qwen3.8-Flash-Next-NVFP4-2x-DGX-Spark/blob/main/README.md)
- decode_tps: 58 (1 request) — [source](https://github.com/Weschera/Qwen3.8-Flash-Next-NVFP4-2x-DGX-Spark/blob/main/README.md)

## Weights

- [nvidia/Qwen3.8-Flash-Next-NVFP4](https://huggingface.co/nvidia/Qwen3.8-Flash-Next-NVFP4) (main, other)

[https://github.com/Weschera/Qwen3.8-Flash-Next-NVFP4-2x-DGX-Spark](https://github.com/Weschera/Qwen3.8-Flash-Next-NVFP4-2x-DGX-Spark)

## Related

- [Local AI on your hardware.](https://www.contextstudios.ai/local-ai.md)