Instructions to use amd/DeepSeek-V4-Flash-EXTENSOR with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Notebooks
Google Colab
Kaggle
Local Apps Settings

How to use amd/DeepSeek-V4-Flash-EXTENSOR 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 amd/DeepSeek-V4-Flash-EXTENSOR:MXFP4
# Run inference directly in the terminal:
llama cli -hf amd/DeepSeek-V4-Flash-EXTENSOR:MXFP4

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf amd/DeepSeek-V4-Flash-EXTENSOR:MXFP4
# Run inference directly in the terminal:
llama cli -hf amd/DeepSeek-V4-Flash-EXTENSOR:MXFP4

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 amd/DeepSeek-V4-Flash-EXTENSOR:MXFP4
# Run inference directly in the terminal:
./llama-cli -hf amd/DeepSeek-V4-Flash-EXTENSOR:MXFP4

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 amd/DeepSeek-V4-Flash-EXTENSOR:MXFP4
# Run inference directly in the terminal:
./build/bin/llama-cli -hf amd/DeepSeek-V4-Flash-EXTENSOR:MXFP4

Use Docker

docker model run hf.co/amd/DeepSeek-V4-Flash-EXTENSOR:MXFP4

LM Studio
Jan

vLLM

How to use amd/DeepSeek-V4-Flash-EXTENSOR with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "amd/DeepSeek-V4-Flash-EXTENSOR"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "amd/DeepSeek-V4-Flash-EXTENSOR",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/amd/DeepSeek-V4-Flash-EXTENSOR:MXFP4

Ollama
How to use amd/DeepSeek-V4-Flash-EXTENSOR with Ollama:
```
ollama run hf.co/amd/DeepSeek-V4-Flash-EXTENSOR:MXFP4
```

Unsloth Studio

How to use amd/DeepSeek-V4-Flash-EXTENSOR 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 amd/DeepSeek-V4-Flash-EXTENSOR 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 amd/DeepSeek-V4-Flash-EXTENSOR to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for amd/DeepSeek-V4-Flash-EXTENSOR to start chatting

Docker Model Runner
How to use amd/DeepSeek-V4-Flash-EXTENSOR with Docker Model Runner:
```
docker model run hf.co/amd/DeepSeek-V4-Flash-EXTENSOR:MXFP4
```

Lemonade

How to use amd/DeepSeek-V4-Flash-EXTENSOR with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull amd/DeepSeek-V4-Flash-EXTENSOR:MXFP4

Run and chat with the model

lemonade run user.DeepSeek-V4-Flash-EXTENSOR-MXFP4

List all available models

lemonade list

Atomic Chat

DeepSeek-V4-Flash-0731 EXTENSOR

An EXTENSOR runtime image derived from deepseek-ai/DeepSeek-V4-Flash-0731. It contains the target model only; the attached DSpark speculative module is not included.

Artifact

File: DeepSeek-V4-Flash-EXTENSOR-MXFP4-Tile16-v2.extensor.gguf
Size: 155,848,011,776 bytes
SHA-256: 9ce000f927b3693b508c523551453cd61b59363ffaef3f1949914d402c0674cd
Quantization: source-native MXFP4 Tile16

Requires EXTENSOR 2.0.0. This file is not compatible with llama.cpp.

Source

deepseek-ai/DeepSeek-V4-Flash-0731 at revision 9e165c30e2704aec5d9d593cce3eebd58bbef1cb.

License

MIT License. See the upstream repository for usage and citation details.

Downloads last month: 2,644

GGUF

Model size

284B params

Architecture

deepseek4

Hardware compatibility

4-bit

Model tree for amd/DeepSeek-V4-Flash-EXTENSOR

Base model

deepseek-ai/DeepSeek-V4-Flash-0731

Quantized

(105)

this model