Instructions to use hampsonw/DeepSeek-V4-Flash-0731-DIQ3S-GIQ1S-UIQ1M-GGUF with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Notebooks
Google Colab
Kaggle
Local Apps Settings

How to use hampsonw/DeepSeek-V4-Flash-0731-DIQ3S-GIQ1S-UIQ1M-GGUF with llama.cpp:

Install (macOS, Linux)

curl -LsSf https://llama.app/install.sh | sh
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf hampsonw/DeepSeek-V4-Flash-0731-DIQ3S-GIQ1S-UIQ1M-GGUF
# Run inference directly in the terminal:
llama cli -hf hampsonw/DeepSeek-V4-Flash-0731-DIQ3S-GIQ1S-UIQ1M-GGUF

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf hampsonw/DeepSeek-V4-Flash-0731-DIQ3S-GIQ1S-UIQ1M-GGUF
# Run inference directly in the terminal:
llama cli -hf hampsonw/DeepSeek-V4-Flash-0731-DIQ3S-GIQ1S-UIQ1M-GGUF

Use pre-built binary

# Download pre-built binary from:
# https://github.com/ggerganov/llama.cpp/releases
# Start a local OpenAI-compatible server with a web UI:
./llama-server -hf hampsonw/DeepSeek-V4-Flash-0731-DIQ3S-GIQ1S-UIQ1M-GGUF
# Run inference directly in the terminal:
./llama-cli -hf hampsonw/DeepSeek-V4-Flash-0731-DIQ3S-GIQ1S-UIQ1M-GGUF

Build from source code

git clone https://github.com/ggerganov/llama.cpp.git
cd llama.cpp
cmake -B build
cmake --build build -j --target llama-server llama-cli
# Start a local OpenAI-compatible server with a web UI:
./build/bin/llama-server -hf hampsonw/DeepSeek-V4-Flash-0731-DIQ3S-GIQ1S-UIQ1M-GGUF
# Run inference directly in the terminal:
./build/bin/llama-cli -hf hampsonw/DeepSeek-V4-Flash-0731-DIQ3S-GIQ1S-UIQ1M-GGUF

Use Docker

docker model run hf.co/hampsonw/DeepSeek-V4-Flash-0731-DIQ3S-GIQ1S-UIQ1M-GGUF

LM Studio
Jan

vLLM

How to use hampsonw/DeepSeek-V4-Flash-0731-DIQ3S-GIQ1S-UIQ1M-GGUF with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "hampsonw/DeepSeek-V4-Flash-0731-DIQ3S-GIQ1S-UIQ1M-GGUF"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "hampsonw/DeepSeek-V4-Flash-0731-DIQ3S-GIQ1S-UIQ1M-GGUF",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/hampsonw/DeepSeek-V4-Flash-0731-DIQ3S-GIQ1S-UIQ1M-GGUF

Ollama
How to use hampsonw/DeepSeek-V4-Flash-0731-DIQ3S-GIQ1S-UIQ1M-GGUF with Ollama:
```
ollama run hf.co/hampsonw/DeepSeek-V4-Flash-0731-DIQ3S-GIQ1S-UIQ1M-GGUF
```

Unsloth Studio

How to use hampsonw/DeepSeek-V4-Flash-0731-DIQ3S-GIQ1S-UIQ1M-GGUF with Unsloth Studio:

Install Unsloth Studio (macOS, Linux, WSL)

curl -fsSL https://unsloth.ai/install.sh | sh
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for hampsonw/DeepSeek-V4-Flash-0731-DIQ3S-GIQ1S-UIQ1M-GGUF to start chatting

Install Unsloth Studio (Windows)

irm https://unsloth.ai/install.ps1 | iex
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for hampsonw/DeepSeek-V4-Flash-0731-DIQ3S-GIQ1S-UIQ1M-GGUF to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for hampsonw/DeepSeek-V4-Flash-0731-DIQ3S-GIQ1S-UIQ1M-GGUF to start chatting

How to use hampsonw/DeepSeek-V4-Flash-0731-DIQ3S-GIQ1S-UIQ1M-GGUF with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf hampsonw/DeepSeek-V4-Flash-0731-DIQ3S-GIQ1S-UIQ1M-GGUF

Configure the model in Pi

# Install Pi:
npm install -g @mariozechner/pi-coding-agent
# Add to ~/.pi/agent/models.json:
{
  "providers": {
    "llama-cpp": {
      "baseUrl": "http://localhost:8080/v1",
      "api": "openai-completions",
      "apiKey": "none",
      "models": [
        {
          "id": "hampsonw/DeepSeek-V4-Flash-0731-DIQ3S-GIQ1S-UIQ1M-GGUF"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

OpenClaw new

How to use hampsonw/DeepSeek-V4-Flash-0731-DIQ3S-GIQ1S-UIQ1M-GGUF with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf hampsonw/DeepSeek-V4-Flash-0731-DIQ3S-GIQ1S-UIQ1M-GGUF

Configure OpenClaw

# Install OpenClaw:
npm install -g openclaw@latest
# Register the local server and set it as the default model:
openclaw onboard --non-interactive --mode local \
  --auth-choice custom-api-key \
  --custom-base-url http://127.0.0.1:8080/v1 \
  --custom-model-id "hampsonw/DeepSeek-V4-Flash-0731-DIQ3S-GIQ1S-UIQ1M-GGUF" \
  --custom-provider-id llama-cpp \
  --custom-compatibility openai \
  --custom-text-input \
  --accept-risk \
  --skip-health

Run OpenClaw

openclaw agent --local --agent main --message "Hello from Hugging Face"

Docker Model Runner
How to use hampsonw/DeepSeek-V4-Flash-0731-DIQ3S-GIQ1S-UIQ1M-GGUF with Docker Model Runner:
```
docker model run hf.co/hampsonw/DeepSeek-V4-Flash-0731-DIQ3S-GIQ1S-UIQ1M-GGUF
```

Lemonade

How to use hampsonw/DeepSeek-V4-Flash-0731-DIQ3S-GIQ1S-UIQ1M-GGUF with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull hampsonw/DeepSeek-V4-Flash-0731-DIQ3S-GIQ1S-UIQ1M-GGUF

Run and chat with the model

lemonade run user.DeepSeek-V4-Flash-0731-DIQ3S-GIQ1S-UIQ1M-GGUF-{{QUANT_TAG}}

List all available models

lemonade list

Hermes Agent

How to use hampsonw/DeepSeek-V4-Flash-0731-DIQ3S-GIQ1S-UIQ1M-GGUF with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf hampsonw/DeepSeek-V4-Flash-0731-DIQ3S-GIQ1S-UIQ1M-GGUF

Configure Hermes

# Install Hermes:
curl -fsSL https://hermes-agent.nousresearch.com/install.sh | bash
hermes setup
# Point Hermes at the local server:
hermes config set model.provider custom
hermes config set model.base_url http://127.0.0.1:8080/v1
hermes config set model.default hampsonw/DeepSeek-V4-Flash-0731-DIQ3S-GIQ1S-UIQ1M-GGUF

Run Hermes

hermes

Atomic Chat

DeepSeek-V4-Flash-0731 role-splice experiment

This is an experimental role-spliced GGUF for deepseek-ai/DeepSeek-V4-Flash-0731. It is not an official DeepSeek release.

The file combines already-quantized tensor payloads from four revision-pinned GGUFs. No tensor was dequantized or requantized during assembly.

Expert recipe

Routed-expert role	Tensors	Type	Donor
`ffn_down_exps.weight`	43	`IQ3_S`	TacoTakumi
`ffn_gate_exps.weight`	43	`IQ1_S`	6block
`ffn_up_exps.weight`	43	`IQ1_M`	AtomicChat

The other 1,199 tensors come from antirez/deepseek-v4-gguf, including its Q8/F16/F32 attention, shared-expert, router, output, and auxiliary tensors.

Complete tensor-type census:

Type	Tensors
`F32`	492
`F16`	359
`Q8_0`	345
`IQ3_S`	43
`IQ1_S`	43
`IQ1_M`	43
`I32`	3

File and verification

File: DeepSeek-V4-Flash-0731-DIQ3S-GIQ1S-UIQ1M.gguf
Size: 86,720,111,200 bytes (80.76 GiB)
Tensors: 1,328
SHA-256: 7cd6c76a910df9a05567e07ad3955e93ec9373fddc2c5732382d9214261593e9
Manifest SHA-256: c0c5c28c4cc48e3845368686c63649e462afb9a07ce597cf4fc306dbb9495de1

splice-manifest.json records every source repository revision, LFS SHA-256, source byte range, tensor type, shape, and output offset. verification.json records structural verification and the runtime smoke test.

The assembled file passed two checks:

A complete parser check matched all 1,328 tensor names, dimensions, types, and offsets against the manifest and calculated the whole-file SHA-256.
CPU-only llama-cli from llama.cpp revision 2f56fc3431f47fe042bf3825e4d5523bdddda993 loaded the model, initialized a 128-token context, evaluated Hello, generated one token, and exited successfully.

Important limitation

This splice has not been quality-benchmarked. The three expert donors used different importance matrices and calibration corpora. GGUF tensor blocks are independently decodable, but combining them does not prove that the resulting model preserves the quality of any donor release.

Treat this as an experimental candidate until it has been compared with the Antirez, Unsloth, and uniform baselines on the same KL/perplexity and task harness.

The file does not include an MTP or DSpark sidecar.

Runtime

Use a llama.cpp build with deepseek4 support. The smoke test used revision 2f56fc3.

llama-cli \
  -m DeepSeek-V4-Flash-0731-DIQ3S-GIQ1S-UIQ1M.gguf \
  -c 4096 \
  -p "Hello"

Configure GPU offload and context size for your hardware. The GGUF alone occupies 80.76 GiB before runtime buffers and KV cache.

Credits and license

The base model and inherited license come from deepseek-ai/DeepSeek-V4-Flash-0731. Quantized payloads are credited to Antirez, TacoTakumi, 6block, and AtomicChat at the pinned revisions above.

License: MIT.

Downloads last month: 276

GGUF

Model size

284B params

Architecture

deepseek4

Hardware compatibility

We're not able to determine the quantization variants.

View all variants

Model tree for hampsonw/DeepSeek-V4-Flash-0731-DIQ3S-GIQ1S-UIQ1M-GGUF

Base model

deepseek-ai/DeepSeek-V4-Flash-0731

Quantized

(127)

this model