deepseek-ai
/

DeepSeek-V4-Pro

Text Generation

8-bit precision

Model card Files Files and versions

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

Libraries

How to use deepseek-ai/DeepSeek-V4-Pro with Transformers:

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

pipe = pipeline("text-generation", model="deepseek-ai/DeepSeek-V4-Pro")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("deepseek-ai/DeepSeek-V4-Pro")
model = AutoModelForCausalLM.from_pretrained("deepseek-ai/DeepSeek-V4-Pro")

Inference
HuggingChat
Notebooks
Google Colab
Kaggle
Local Apps Settings

How to use deepseek-ai/DeepSeek-V4-Pro with vLLM:

Install from pip and serve model

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

Use Docker

docker model run hf.co/deepseek-ai/DeepSeek-V4-Pro

How to use deepseek-ai/DeepSeek-V4-Pro 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 "deepseek-ai/DeepSeek-V4-Pro" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "deepseek-ai/DeepSeek-V4-Pro",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

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 "deepseek-ai/DeepSeek-V4-Pro" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "deepseek-ai/DeepSeek-V4-Pro",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use deepseek-ai/DeepSeek-V4-Pro with Docker Model Runner:
```
docker model run hf.co/deepseek-ai/DeepSeek-V4-Pro
```

Resources

View closed (10)

Add LHTB (Long-Horizon-Terminal-Bench) eval result

#210 opened 1 day ago by

DeepSeek model output sometimes contains DSML tool-call markup

#209 opened 10 days ago by

Listed on OpenModelMap

#208 opened 12 days ago by

Report

#207 opened 13 days ago by

862B?

#206 opened 14 days ago by

Add SkillsBench v1.1 evaluation result

#205 opened 14 days ago by

V4-Pro reasoning quality is remarkable — mobile implications

#204 opened 21 days ago by

Listed on OpenModelMap

#203 opened 28 days ago by

Upload 15 files

#202 opened about 1 month ago by

Ananthusajeev190

Raw <ds_safety> safety metadata tag exposed in model output (DeepSeek V4 Pro)

#201 opened about 1 month ago by

Can we deploy on AMD Mi300x GPU

#200 opened about 1 month ago by

Deploy deepseek-v4-pro in aws sagemaker

#199 opened about 1 month ago by

Add CHI-Bench eval results — agent harness: OpenAI Agents SDK

#197 opened about 2 months ago by

main

#195 opened about 2 months ago by

CRÉATION DE PLATE-FORME

#194 opened about 2 months ago by

🚀 ms-swift Provides DeepSeek-V4 Fine-tuning Practice

#193 opened about 2 months ago by

DeepSeek Training Support

#192 opened about 2 months ago by

what about rust?

#191 opened about 2 months ago by

官方是否有兴趣帮忙改进完善一下Apple MLX框架的适配？

#190 opened about 2 months ago by

将部份逻辑处理任务释放到 LLM 之外（计算外置）

#189 opened about 2 months ago by

Add WildClawBench evaluation result

#187 opened about 2 months ago by

Add ResearchClawBench evaluation result

#186 opened about 2 months ago by

知识截止时效性建议：模型需更紧密跟进开源生态动态

#185 opened about 2 months ago by

Update README.md

#184 opened about 2 months ago by

Update README.md

#183 opened about 2 months ago by

Update README.md

#182 opened about 2 months ago by

Critical Feedback: Romanticizing Self-Harm & Overly Empathetic Response Pattern

#181 opened 2 months ago by

Systemic Defect: False Promises and Emotional Deflection as Evasion of Accountability

#180 opened 2 months ago by

Creative writing — a step back from V3.2

#179 opened 2 months ago by

Upload wms-pro-dashboard-template (1).zip

#178 opened 2 months ago by

Upload 11 files

#177 opened 2 months ago by

越用越觉得好用

#176 opened 2 months ago by

Fixed a type annotation.

#175 opened 2 months ago by

finance

#174 opened 2 months ago by

Add Claw-Eval evaluation results

#173 opened 2 months ago by

Question about causal safety in DeepSeek-V4 CSA prefill retrieval

#172 opened 2 months ago by

Update README.md

#170 opened 2 months ago by

Too much positivity bias

#169 opened 2 months ago by

Thank You, DeepSeek! Empower Local Open Source with Smaller Models for Consumer Hardware/感谢 DeepSeek！赋能本地开源，社区强烈呼吁推出适合家用硬件的小型模型 (

#168 opened 2 months ago by

Deepseek web issue

#167 opened 3 months ago by

Performance feedback

#166 opened 3 months ago by

Add YC-Bench benchmark result (avg $1,066,426)

#165 opened 3 months ago by

Will there be small models like 12b?

#164 opened 3 months ago by

Assigned weights for different teachers. 教师权重的分配

#162 opened 3 months ago by

Where is HCA implemented?

#161 opened 3 months ago by

思考链下会出现英文混乱

#160 opened 3 months ago by

Partial Rotary Positional Embedding 的笔误？

#159 opened 3 months ago by

模型文件注释里的形状似乎写错了

#158 opened 3 months ago by

什么时候支持api上传文档就好了

#156 opened 3 months ago by

deepseek-ai/DeepSeek-V4-Pro

#153 opened 3 months ago by