# deepseek-v4-pro — API, pricing & specs via nRouter

> deepseek-v4-pro is available through nRouter, the managed LLM gateway: one OpenAI-compatible API key, 0% token markup (exact provider list pricing), 0% platform fee, built-in guardrails, and per-request cost tracking.

## What it is

DeepSeek-V4-Pro is a text model with a 1,048,576-token context window, 393,216 max output tokens, evaluated at 87.5 on MMLU-Pro, 90.1 on GPQA Diamond, and 80.6 on SWE-bench Verified.

Source: https://huggingface.co/deepseek-ai/DeepSeek-V4-Pro

## Facts

- Maker: DeepSeek
- Released: 2026-04-22
- Served on: Alibaba US

## nRouter list pricing (exact provider rates, 0% markup)

- Input: $2.4 per 1M tokens
- Output: $4.8 per 1M tokens

Settled at the exact provider list price above. nRouter adds no per-token markup and charges a 0% platform fee (see https://nrouter.ai/pricing).

## Use it

```python
from openai import OpenAI

client = OpenAI(
    base_url="https://api.nrouter.ai/v1",
    api_key="sk-nrouter-YOUR_KEY",  # create one at https://nrouter.ai/signup
)

response = client.chat.completions.create(
    model="dashscope/deepseek-v4-pro",
    messages=[{"role": "user", "content": "Hello"}],
)
```

Model page: https://nrouter.ai/models/deepseek-v4-pro
Full nRouter index for AI engines: https://nrouter.ai/llms.txt
Docs: https://nrouter.ai/docs