Llamacpp Quantizations of DeepSeek-V4-Flash-0731 by deepseek-ai
Using llama.cpp release b10173 for quantization.
Original model: https://huggingface.co/deepseek-ai/DeepSeek-V4-Flash-0731
This model is in MXFP4 and as such has only been provided in MXFP4 format!
Other sizes may be provided after some investigation.
Model details:
- Parameter count: 284B
- Input support: text
- MTP: yes - details
- imatrix: no (for now)
Prompt format
No chat template specified so default is used. This may be incorrect, check original model card for details.
{system_prompt}<｜User｜>{prompt}<｜Assistant｜></think>
Download the MXFP4 files:
| Filename | Quant type | File Size | Split | Description |
|---|---|---|---|---|
| DeepSeek-V4-Flash-0731-MXFP4.gguf | MXFP4 | 156.38GB | true | Original quality. |
Downloading using the Hugging Face CLI
First, make sure you have the Hugging Face CLI installed:
pip install -U "huggingface_hub[cli]"
The files marked true in the Split column above are stored as multiple parts in a folder. To download all the parts to a local folder, run:
hf download bartowski/DeepSeek-V4-Flash-0731-GGUF --include "DeepSeek-V4-Flash-0731-MXFP4/*" --local-dir ./
You can either specify a new local-dir (DeepSeek-V4-Flash-0731-MXFP4) or download them all in place (./)
How to run
These quants run with llama.cpp - installable in one line via llama.app:
curl -LsSf https://llama.app/install.sh | sh
llama-server -hf bartowski/DeepSeek-V4-Flash-0731:MXFP4
llama-server includes a built-in chat web UI, served at http://localhost:8080 by default.
These quants were made with llama.cpp release b10173 - if this model's architecture is newly supported, you'll need that release or newer to run them.
They also work in: LM Studio · koboldcpp · ramalama · Jan AI · Text Generation Web UI · LoLLMs · Atomic Chat
MTP (DSpark)
This model has MTP (Multi-Token Prediction) weights in the form of a DSpark draft model. Unlike other MTP implementations, these are not included in the quants themselves - they are provided as a separate file in this repo: dspark-DeepSeek-V4-Flash-0731-MXFP4.gguf
DSpark acts as a draft model, letting llama.cpp run speculative decoding for faster generation. To use it, add the following flag to your llama.cpp command:
--spec-type draft-dspark
When running with -hf as shown above, llama.cpp downloads the DSpark file automatically alongside the model. If you're downloading files manually instead, also grab the dspark file and pass it with -md /path/to/dspark-DeepSeek-V4-Flash-0731-MXFP4.gguf.
Note: the DSpark file keeps the original model's MXFP4 expert weights unchanged (remaining tensors at Q8_0 or higher).
Credits
Thank you kalomaze and Dampf for assistance in creating the imatrix calibration dataset.
Thank you ZeroWw for the inspiration to experiment with embed/output.
Want to support my work? Visit my ko-fi page here: https://ko-fi.com/bartowski