---
type: "Article"
title: "DeepSeek-V4-Flash 284B IQ2_XXS with DwarfStar on 1× Mac (vision)"
description: "DeepSeek-V4-Flash 284B IQ2_XXS with DwarfStar on 1× Mac (vision): recipe by Weschera with speed, condition and source for every number, weights and license."
resource: "https://www.contextstudios.ai/local-ai/weschera--dsv4-flash-vision-exp-iq2xxs-dspark-ds4-mac-studio-m4-max"
language: "en"
generated:
  by: "process:contextstudios-md/1"
  at: "2026-10-02T21:54:34.992Z"
status: "stable"
---

# DeepSeek-V4-Flash 284B IQ2_XXS with DwarfStar on 1× Mac (vision)

DeepSeek-V4-Flash 284B IQ2_XXS with DwarfStar on 1× Mac (vision): 28.5 tok/s according to github.com (dataset as of Sep 29, 2026).

- Creator: [Weschera](https://www.contextstudios.ai/local-ai/creators/weschera.md)
- Engine: ds4 (DwarfStar)
- Quantization: IQ2_XXS imatrix q2 (antirez), w2Q2K/AProjQ8/SExpQ8/OutQ8; GGUF 86,720,111,776 bytes
- Model family: DeepSeek-V4-Flash
- Hardware: mac x1
- Context: 1048576

## All numbers

- 28.5 tok/s (Everyday; Everyday, 1 request, prose prompt, context 32k, with DSpark (per recipe)) — [source](https://github.com/Weschera/DeepSeek-V4-Flash-Vision-Exp-ds4-Mac-Studio/blob/main/README.md)
- decode_tps: 28.5 (1 request) — [source](https://github.com/Weschera/DeepSeek-V4-Flash-Vision-Exp-ds4-Mac-Studio/blob/main/README.md)
- decode_tps: 31.7 (1 request) — [source](https://github.com/Weschera/DeepSeek-V4-Flash-Vision-Exp-ds4-Mac-Studio/blob/main/README.md)

## Weights

- [antirez/deepseek-v4-gguf](https://huggingface.co/antirez/deepseek-v4-gguf) (main, mit)
- [antirez/deepseek-v4-gguf](https://huggingface.co/antirez/deepseek-v4-gguf) (draft, mit)
- [antirez/deepseek-v4-gguf](https://huggingface.co/antirez/deepseek-v4-gguf) (vision, mit)

[https://github.com/Weschera/DeepSeek-V4-Flash-Vision-Exp-ds4-Mac-Studio](https://github.com/Weschera/DeepSeek-V4-Flash-Vision-Exp-ds4-Mac-Studio)

## Related

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