DeepSeek V4 Pro API Guide

DeepSeek V4 Pro is one of the strongest open-weight models from China, rivaling GPT-5 and Claude in reasoning benchmarks. On Qubax, it's available at 49% off OpenRouter — just $0.22/1M input tokens.

Why DeepSeek V4 Pro?

• Strong reasoning: Competes with Claude Sonnet 5 on math and code benchmarks
• Large context: 1M token context window
• Open weight: You can verify the model isn't censored or modified
• Cheap: $0.22/1M input on Qubax vs $0.43/1M on OpenRouter

Quick Start

from openai import OpenAI

client = OpenAI(
    base_url="https://api.qubax.ai/v1",
    api_key="qbx_live_YOUR_KEY",
)

response = client.chat.completions.create(
    model="deepseek-v4-pro",
    messages=[
        {"role": "user", "content": "Solve: A train travels 300km in 4 hours. What's the average speed?"}
    ],
)

print(response.choices[0].message.content)

cURL Example

curl https://api.qubax.ai/v1/chat/completions \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer qbx_live_YOUR_KEY" \
  -d '{
    "model": "deepseek-v4-pro",
    "messages": [{"role": "user", "content": "Write a Rust HTTP server"}]
  }'

Use with Cursor IDE

1. Open Cursor Settings → Models
2. Set Override OpenAI Base URL to https://api.qubax.ai/v1
3. Enter your Qubax API key
4. Set model to deepseek-v4-pro

Cost for Common Workloads

Related Models

• DeepSeek V4 Pro API
• GLM 5.2 — Another top Chinese LLM, 93% off
• Qwen 3 235B — Alibaba's flagship
• All models

Try DeepSeek V4 Pro at 49% off — [get $1 free credits](https://qubax.ai/register).