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

Libraries

How to use unsloth/DeepSeek-V4-Flash with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("text-generation", model="unsloth/DeepSeek-V4-Flash")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("unsloth/DeepSeek-V4-Flash")
model = AutoModelForCausalLM.from_pretrained("unsloth/DeepSeek-V4-Flash")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use unsloth/DeepSeek-V4-Flash with vLLM:

Install from pip and serve model

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

Use Docker

docker model run hf.co/unsloth/DeepSeek-V4-Flash

SGLang

How to use unsloth/DeepSeek-V4-Flash with SGLang:

Install from pip and serve model

# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
    --model-path "unsloth/DeepSeek-V4-Flash" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "unsloth/DeepSeek-V4-Flash",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker images

docker run --gpus all \
    --shm-size 32g \
    -p 30000:30000 \
    -v ~/.cache/huggingface:/root/.cache/huggingface \
    --env "HF_TOKEN=<secret>" \
    --ipc=host \
    lmsysorg/sglang:latest \
    python3 -m sglang.launch_server \
        --model-path "unsloth/DeepSeek-V4-Flash" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "unsloth/DeepSeek-V4-Flash",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

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

waiting for gguf version

by lvyufeng - opened 13 days ago

Discussion

lvyufeng

13 days ago

waiting for gguf version

thomasmaindron

12 days ago

does anyone know what actually changes in the safetensors versions unsloth publishes compared to the original ones?

jesleocizi

12 days ago

does anyone know what actually changes in the safetensors versions unsloth publishes compared to the original ones?

Just a mirror I think

jesleocizi

12 days ago

waiting for gguf version

Still very fresh model, not integrated into many frameworks so we need to wait. There's not jinja in this release unfortunately

ait3st3r

12 days ago

No 4 bit?

asidaddy

12 days ago

guys, this isn't supported by llamacpp yet, it a hassle to run with it as well.

jesleocizi

12 days ago

guys, this isn't supported by llamacpp yet, it a hassle to run with it as well.

We all already waiting for that. Just need few days I believe, and some decent GGUF, but unsloth team always provides https://github.com/ggml-org/llama.cpp/issues/22319

TeaDiffusion

10 days ago

guys, this isn't supported by llamacpp yet, it a hassle to run with it as well.

is the 140W - 160W range the max power draw while running this model (off-loaded to the two GPUs)?

Upload images, audio, and videos by dragging in the text input, pasting, or clicking here.

Tap or paste here to upload images

· Sign up or log in to comment