---
type: "Article"
title: "DeepSeek-V4-Flash 291B FP8 with vLLM on 4× DGX Spark"
description: "DeepSeek-V4-Flash 291B FP8 with vLLM on 4× DGX Spark: recipe by tonyd2wild with speed, condition and source for every number, weights and license."
resource: "https://www.contextstudios.ai/local-ai/tonyd2wild--deepseek-v4-flash-tp4-4x-dgx-spark"
language: "en"
generated:
  by: "process:contextstudios-md/1"
  at: "2026-10-02T21:54:59.332Z"
status: "stable"
---

# DeepSeek-V4-Flash 291B FP8 with vLLM on 4× DGX Spark

DeepSeek-V4-Flash 291B FP8 with vLLM on 4× DGX Spark: 46 tok/s according to github.com (dataset as of Sep 29, 2026).

- Creator: [tonyd2wild](https://www.contextstudios.ai/local-ai/creators/tonyd2wild.md)
- Engine: vLLM
- Quantization: FP8
- Model family: DeepSeek-V4-Flash
- Hardware: dgx-spark x4

## All numbers

- 46 tok/s (Everyday; Everyday, 1 request, prose prompt, with DSpark) — [source](https://github.com/tonyd2wild/Deepseek-V4-Flash-TP4-4x-DGX-Spark/blob/main/README.md)
- decode_tps: 46 (1 request) — [source](https://github.com/tonyd2wild/Deepseek-V4-Flash-TP4-4x-DGX-Spark/blob/main/README.md)
- decode_tps: 120 (1 request) — [source](https://github.com/tonyd2wild/Deepseek-V4-Flash-TP4-4x-DGX-Spark/blob/main/README.md)

## Weights

- [deepseek-ai/DeepSeek-V4-Flash](https://huggingface.co/deepseek-ai/DeepSeek-V4-Flash) (main, mit)

[https://github.com/tonyd2wild/Deepseek-V4-Flash-TP4-4x-DGX-Spark](https://github.com/tonyd2wild/Deepseek-V4-Flash-TP4-4x-DGX-Spark)

## Related

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