DeepSeek V4 Pro

DeepSeek V4 Pro is the latest flagship model from DeepSeek. Leading performance on coding, reasoning, and multilingual tasks. 128K context window handles entire codebases and long documents.

Use Cases

Quick Start

curl https://aiwave.live/v1/chat/completions \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -d '{"model":"deepseek-v4-pro","messages":[{"role":"user","content":"Hello!"}]}'

What it costs in practice

These figures come straight from the live rate above — monthly token volume multiplied by the per-million price. Adjust the volumes to match your own traffic.

Workload	Assumption	Monthly tokens	Monthly cost
Chatbot	~50k short conversations	5.0M in / 1.5M out	$8.70
Coding assistant	one developer, full-time	20.0M in / 4.0M out	$30.45
RAG / document Q&A	long retrieved contexts	50.0M in / 2.0M out	$58.72

Rates are pulled from the live pricing endpoint. Token accounting is returned in the usage field of every non-streaming response.

Quick start

The endpoint is OpenAI-compatible, so the official SDKs work once you change base_url. Nothing else in your code needs to move.

from openai import OpenAI

client = OpenAI(
    api_key="sk-YOUR_KEY",
    base_url="https://aiwave.live/v1",
)

response = client.chat.completions.create(
    model="deepseek-v4-pro",
    messages=[{"role": "user", "content": "Hello"}],
)
print(response.choices[0].message.content)
print(response.usage)   # exact token counts for cost tracking

Streaming uses the same server-sent-event format as OpenAI:

stream = client.chat.completions.create(
    model="deepseek-v4-pro",
    messages=[{"role": "user", "content": "Explain vector search"}],
    stream=True,
)
for chunk in stream:
    delta = chunk.choices[0].delta.content
    if delta:
        print(delta, end="", flush=True)

Node.js, LangChain, LlamaIndex, the Vercel AI SDK, Cursor and Cline all accept a custom base URL, so they work with this model without a dedicated provider plugin. Full parameter reference lives in the Chat Completions docs.

How it compares to sibling models

Same provider, same API surface — the practical difference is price and capability. Because every model here speaks the OpenAI wire format, switching between them is a one-word change to the model field.

Model ID	Input / 1M	Output / 1M
deepseek-r1-distill-qwen-14b	$0.0154	$0.0308
deepseek-r1-distill-qwen-32b	$0.0154	$0.0308
deepseek-v3	$0.154	$0.308
deepseek-v3.2-think	$0.154	$0.308
deepseek-chat	$0.182	$0.364
deepseek-v4-flash	$0.206	$0.412

Where to go next

• API documentation — auth, endpoints, error codes
• Live pricing for every model
• Model directory — compare context windows and capabilities
• Access Chinese AI models without a Chinese phone number
• Migrate an OpenAI app in two lines

Related Models