GraySoft
Projects Models Compare Cloud benchmarks FAQ Download guIDE →
Model Intelligence Sheet

tarruda/DeepSeek-V4-Flash-GGUF overview

DeepSeek V4 Flash GGUF GGUF quantizations for deepseek ai/DeepSeek V4 Flash https://huggingface.co/deepseek ai/DeepSeek V4 Flash DeepSeek published the origina…

ggufimatrixconversationaldeepseekllama.cpptext-generationbase_model:deepseek-ai/DeepSeek-V4-Flashbase_model:quantized:deepseek-ai/DeepSeek-V4-Flashlicense:mitendpoints_compatibleregion:us

Runs locally from ~5.0 MB disk (4 GB VRAM class GPUs with llama.cpp / guIDE).

Downloads
0
Likes
10
Pipeline
text-generation
Author

Repository Files & Downloads

17 GGUF files detected
Direct downloads for local inference
FileTypeQuantizationSizeLink
IQ3_XXS/DeepSeek-V4-Flash-IQ3_XXS-00001-of-00004.ggufGGUFIQ3_XXS5.0 MBDownload
IQ3_XXS/DeepSeek-V4-Flash-IQ3_XXS-00002-of-00004.ggufGGUFIQ3_XXS45.80 GBDownload
IQ3_XXS/DeepSeek-V4-Flash-IQ3_XXS-00003-of-00004.ggufGGUFIQ3_XXS46.56 GBDownload
IQ3_XXS/DeepSeek-V4-Flash-IQ3_XXS-00004-of-00004.ggufGGUFIQ3_XXS12.04 GBDownload
MXFP4/DeepSeek-V4-Flash-MXFP4-00001-of-00005.ggufGGUFGGUF5.0 MBDownload
MXFP4/DeepSeek-V4-Flash-MXFP4-00002-of-00005.ggufGGUFGGUF45.70 GBDownload
MXFP4/DeepSeek-V4-Flash-MXFP4-00003-of-00005.ggufGGUFGGUF45.69 GBDownload
MXFP4/DeepSeek-V4-Flash-MXFP4-00004-of-00005.ggufGGUFGGUF45.69 GBDownload
MXFP4/DeepSeek-V4-Flash-MXFP4-00005-of-00005.ggufGGUFGGUF8.56 GBDownload
Q2_K/DeepSeek-V4-Flash-Q2_K-00001-of-00003.ggufGGUFQ2_K5.0 MBDownload
Q2_K/DeepSeek-V4-Flash-Q2_K-00002-of-00003.ggufGGUFQ2_K45.92 GBDownload
Q2_K/DeepSeek-V4-Flash-Q2_K-00003-of-00003.ggufGGUFQ2_K44.37 GBDownload
Q3_K/DeepSeek-V4-Flash-Q3_K-00001-of-00004.ggufGGUFQ3_K5.0 MBDownload
Q3_K/DeepSeek-V4-Flash-Q3_K-00002-of-00004.ggufGGUFQ3_K46.55 GBDownload
Q3_K/DeepSeek-V4-Flash-Q3_K-00003-of-00004.ggufGGUFQ3_K45.82 GBDownload
Q3_K/DeepSeek-V4-Flash-Q3_K-00004-of-00004.ggufGGUFQ3_K24.13 GBDownload
imatrix.ggufGGUFGGUF448.6 MBDownload

Model Details

Model IDtarruda/DeepSeek-V4-Flash-GGUF
Authortarruda
Pipelinetext-generation
Licensemit
Base modeldeepseek-ai/DeepSeek-V4-Flash
Last modified2026-07-02T05:07:00.000Z

Model README

---

quantized_by: tarruda

pipeline_tag: text-generation

base_model: deepseek-ai/DeepSeek-V4-Flash

base_model_relation: quantized

license: mit

license_link: https://huggingface.co/datasets/choosealicense/licenses/blob/main/markdown/mit.md

tags:

  • gguf
  • imatrix
  • conversational
  • deepseek
  • llama.cpp

---

DeepSeek V4 Flash GGUF

GGUF quantizations for

deepseek-ai/DeepSeek-V4-Flash

DeepSeek published the original model weights in MXFP4, so the MXFP4 GGUFs in

this repo are direct conversions of those original safetensors.

Quant Recipes

| Recipe | Quant Size | Default type | Tensor-specific overrides |

| --- | --- | --- | --- |

| Q3_K | 119297.23 MiB (3.52 BPW) | Q6_K | ffn_down_exps=q3_k, ffn_gate_exps=q3_k, ffn_up_exps=q3_k |

| IQ3_XXS | 106912.23 MiB (3.15 BPW) | Q6_K | ffn_down_exps=iq3_xxs, ffn_gate_exps=iq3_xxs, ffn_up_exps=iq3_xxs |

| Q2_K | 92465.23 MiB (2.73 BPW) | Q6_K | ffn_down_exps=q2_k, ffn_gate_exps=q2_k, ffn_up_exps=q2_k |

Usage

This is the script I use to run:

#!/bin/sh -e

model="./IQ3_XXS/DeepSeek-V4-Flash-IQ3_XXS-00001-of-00004.gguf"

ctx=131072
parallel=1

ctx_size=$((ctx * parallel))

llama-server --no-mmap --no-warmup \
  --model $model --ctx-size $ctx_size -np $parallel \
  --temp 1.0 --top-p 1.0

Run tarruda/DeepSeek-V4-Flash-GGUF with guIDE

Download guIDE — the AI-native code editor with local LLM inference and 69 built-in tools.

Download guIDE → · Browse 524k+ models · Compare models

Source: Hugging Face · Compare models