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

Libraries

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

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

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

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

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

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

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

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "MuVeraAI/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": "MuVeraAI/DeepSeek-V4-Flash",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

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

SGLang

How to use MuVeraAI/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 "MuVeraAI/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": "MuVeraAI/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 "MuVeraAI/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": "MuVeraAI/DeepSeek-V4-Flash",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

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

DeepSeek-V4-Flash

File size: 951 Bytes

ed3a5f4

# Inference code for DeepSeek models

First convert huggingface model weight files to the format of this project.
```bash
export EXPERTS=256
export MP=4
export CONFIG=config.json
python convert.py --hf-ckpt-path ${HF_CKPT_PATH} --save-path ${SAVE_PATH} --n-experts ${EXPERTS} --model-parallel ${MP}
```

Then chat with DeepSeek model at will!
```bash
torchrun --nproc-per-node ${MP} generate.py --ckpt-path ${SAVE_PATH} --config ${CONFIG} --interactive
```

Or batch inference from file.
```bash
torchrun --nproc-per-node ${MP} generate.py --ckpt-path ${SAVE_PATH} --config ${CONFIG} --input-file ${FILE}
```

Or multi nodes inference.
```bash
torchrun --nnodes ${NODES} --nproc-per-node $((MP / NODES)) --node-rank $RANK --master-addr $ADDR generate.py --ckpt-path ${SAVE_PATH} --config ${CONFIG} --input-file ${FILE}
```

If you want to use fp8, just remove `"expert_dtype": "fp4"` in `config.json` and specify `--expert-dtype fp8` in `convert.py`.