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

mondk/DeepSeek-V4.1-Flash-1017-GGUF overview

DeepSeek V4.1 Flash 1017 GGUF GGUF quants of deepseek ai/DeepSeek V4.1 Flash 1017 https://huggingface.co/deepseek ai/DeepSeek V4.1 Flash 1017 . | Folder | Size…

ggufllama.cppdeepseekimage-text-to-textbase_model:deepseek-ai/DeepSeek-V4.1-Flash-1017base_model:quantized:deepseek-ai/DeepSeek-V4.1-Flash-1017license:mitendpoints_compatibleregion:usconversational

Runs locally from ~3.74 GB disk (4 GB VRAM class GPUs with llama.cpp / guIDE).

Downloads
0
Likes
0
Pipeline
image-text-to-text
Author

Repository Files & Downloads

30 GGUF files detected
Direct downloads for local inference
FileTypeQuantizationSizeLink
Q3_K_M/DeepSeek-V4.1-Flash-1017-Q3_K_M-00001-of-00010.ggufGGUFQ3_K_M37.05 GBDownload
Q3_K_M/DeepSeek-V4.1-Flash-1017-Q3_K_M-00002-of-00010.ggufGGUFQ3_K_M35.99 GBDownload
Q3_K_M/DeepSeek-V4.1-Flash-1017-Q3_K_M-00003-of-00010.ggufGGUFQ3_K_M35.99 GBDownload
Q3_K_M/DeepSeek-V4.1-Flash-1017-Q3_K_M-00004-of-00010.ggufGGUFQ3_K_M35.99 GBDownload
Q3_K_M/DeepSeek-V4.1-Flash-1017-Q3_K_M-00005-of-00010.ggufGGUFQ3_K_M35.99 GBDownload
Q3_K_M/DeepSeek-V4.1-Flash-1017-Q3_K_M-00006-of-00010.ggufGGUFQ3_K_M35.99 GBDownload
Q3_K_M/DeepSeek-V4.1-Flash-1017-Q3_K_M-00007-of-00010.ggufGGUFQ3_K_M23.99 GBDownload
Q3_K_M/DeepSeek-V4.1-Flash-1017-Q3_K_M-00008-of-00010.ggufGGUFQ3_K_M39.34 GBDownload
Q3_K_M/DeepSeek-V4.1-Flash-1017-Q3_K_M-00009-of-00010.ggufGGUFQ3_K_M39.34 GBDownload
Q3_K_M/DeepSeek-V4.1-Flash-1017-Q3_K_M-00010-of-00010.ggufGGUFQ3_K_M3.74 GBDownload
Q4_K_M/DeepSeek-V4.1-Flash-1017-Q4_K_M-00001-of-00010.ggufGGUFQ4_K_M48.16 GBDownload
Q4_K_M/DeepSeek-V4.1-Flash-1017-Q4_K_M-00002-of-00010.ggufGGUFQ4_K_M44.89 GBDownload
Q4_K_M/DeepSeek-V4.1-Flash-1017-Q4_K_M-00003-of-00010.ggufGGUFQ4_K_M44.89 GBDownload
Q4_K_M/DeepSeek-V4.1-Flash-1017-Q4_K_M-00004-of-00010.ggufGGUFQ4_K_M44.89 GBDownload
Q4_K_M/DeepSeek-V4.1-Flash-1017-Q4_K_M-00005-of-00010.ggufGGUFQ4_K_M44.89 GBDownload
Q4_K_M/DeepSeek-V4.1-Flash-1017-Q4_K_M-00006-of-00010.ggufGGUFQ4_K_M45.98 GBDownload
Q4_K_M/DeepSeek-V4.1-Flash-1017-Q4_K_M-00007-of-00010.ggufGGUFQ4_K_M32.83 GBDownload
Q4_K_M/DeepSeek-V4.1-Flash-1017-Q4_K_M-00008-of-00010.ggufGGUFQ4_K_M51.50 GBDownload
Q4_K_M/DeepSeek-V4.1-Flash-1017-Q4_K_M-00009-of-00010.ggufGGUFQ4_K_M51.50 GBDownload
Q4_K_M/DeepSeek-V4.1-Flash-1017-Q4_K_M-00010-of-00010.ggufGGUFQ4_K_M4.67 GBDownload
Q8_0/DeepSeek-V4.1-Flash-1017-Q8_0-00001-of-00010.ggufGGUFQ8_040.35 GBDownload
Q8_0/DeepSeek-V4.1-Flash-1017-Q8_0-00002-of-00010.ggufGGUFQ8_040.34 GBDownload
Q8_0/DeepSeek-V4.1-Flash-1017-Q8_0-00003-of-00010.ggufGGUFQ8_040.34 GBDownload
Q8_0/DeepSeek-V4.1-Flash-1017-Q8_0-00004-of-00010.ggufGGUFQ8_040.34 GBDownload
Q8_0/DeepSeek-V4.1-Flash-1017-Q8_0-00005-of-00010.ggufGGUFQ8_040.34 GBDownload
Q8_0/DeepSeek-V4.1-Flash-1017-Q8_0-00006-of-00010.ggufGGUFQ8_040.34 GBDownload
Q8_0/DeepSeek-V4.1-Flash-1017-Q8_0-00007-of-00010.ggufGGUFQ8_026.89 GBDownload
Q8_0/DeepSeek-V4.1-Flash-1017-Q8_0-00008-of-00010.ggufGGUFQ8_097.28 GBDownload
Q8_0/DeepSeek-V4.1-Flash-1017-Q8_0-00009-of-00010.ggufGGUFQ8_097.28 GBDownload
Q8_0/DeepSeek-V4.1-Flash-1017-Q8_0-00010-of-00010.ggufGGUFQ8_09.56 GBDownload

Model Details

Model IDmondk/DeepSeek-V4.1-Flash-1017-GGUF
Authormondk
Pipelineimage-text-to-text
Licensemit
Base modeldeepseek-ai/DeepSeek-V4.1-Flash-1017
Last modified2026-10-17T15:16:05.000Z

Model README

---

license: mit

base_model: deepseek-ai/DeepSeek-V4.1-Flash-1017

base_model_relation: quantized

pipeline_tag: image-text-to-text

tags:

  • gguf
  • llama.cpp
  • deepseek

---

DeepSeek-V4.1-Flash-1017 GGUF

GGUF quants of deepseek-ai/DeepSeek-V4.1-Flash-1017.

| Folder | Size | Files | Notes |

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

| Q8_0 | 473 GiB | 10 | Q8_0 everywhere except the routed experts, which keep their native MXFP4. Closest to the original weights. |

| Q4_K_M | 414 GiB | 10 | The routed experts are already about 4-bit, so this saves less than usual. |

| Q3_K_M | 323 GiB | 10 | |

| Q2_K | 246 GiB | 10 | Smallest usable size. |

These are plain quants with no imatrix. Each folder holds a split GGUF. Point llama.cpp at the -00001-of- file and it loads the rest.

Running

Mainline llama.cpp doesn't support DeepSeek V4.1 yet. These files were made with the branch from ggml-org/llama.cpp#28696 at commit 2076412, and they need that branch or a build that includes it.

git clone https://github.com/vcruz305/llama.cpp && cd llama.cpp
git checkout 2076412598e99127ea2554467b2b7530856895d3
cmake -B build && cmake --build build -j

The two engram tables are large and read a few rows per token. Keeping them on the CPU works well, for example -ot "engram_embd.weight=CPU". The PR has a tested command line.

Text only for now. The llama.cpp branch can't convert the vision tower yet.

Acknowledgements

Huge thanks to Hugging Face for generously sponsoring the compute used for this project 🤗

x.com/realmrfakename

Run mondk/DeepSeek-V4.1-Flash-1017-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