DeepSeek
DeepSeek Text

deepseek-v4-pro

Model ID: deepseek-v4-pro
Family
DeepSeek
Context
1,000,000
Minimum charge / request
0.00023 USD
Cache hit rate
84.5%
Requests
61.2K
Success
99.6%
Tokens
6.47B
Uptime 99.49%
24h agoNow
7 days agoNow

Pricing

Input / 1M tokens
0.153 USD
Output / 1M tokens
0.307 USD
Minimum charge / request
0.00023 USD
Prompt Cache 84.5%
Cache read / 1M tokens
0.0115 USD
Cache write / 1M tokens
0.023 USD

Repeated prompts are cached — reads are far cheaper than input, applied automatically via the API.

Cost estimator
Estimated total 0 USD
Average / request 0 USD

Estimate based on list prices. Caching usually makes it cheaper.

Supported endpoints
https://api.xah.io/v1/chat/completions
https://api.xah.io/v1/messages
https://api.xah.io/v1/responses
https://api.xah.io/v1beta/models/deepseek-v4-pro:generateContent
https://api.xah.io/api/chat
Method POST · application/json
Auth Bearer token (API key)
Tested load 🟢 Enterprise peak 69/s · sustained 13.6/s
Latency p95 6,334 ms
API kinds Text

Share

curl 'https://api.xah.io/v1/chat/completions' \
  -H 'Authorization: Bearer YOUR_API_KEY' \
  -H 'Content-Type: application/json' \
  -d '{
    "model": "deepseek-v4-pro",
    "messages": [
        {
            "role": "user",
            "content": "Xin chào, bạn là ai?"
        }
    ]
}'
// Node.js / Browser
const res = await fetch('https://api.xah.io/v1/chat/completions', {
  method: 'POST',
  headers: {
    'Authorization': 'Bearer YOUR_API_KEY',
    'Content-Type': 'application/json'
  },
  body: JSON.stringify({
    "model": "deepseek-v4-pro",
    "messages": [
        {
            "role": "user",
            "content": "Xin chào, bạn là ai?"
        }
    ]
})
});
const data = await res.json();
console.log(data);
import requests

url = 'https://api.xah.io/v1/chat/completions'
headers = {
    'Authorization': 'Bearer YOUR_API_KEY',
    'Content-Type': 'application/json',
}
payload = {
    "model": "deepseek-v4-pro",
    "messages": [
        {
            "role": "user",
            "content": "Xin chào, bạn là ai?"
        }
    ]
}

res = requests.post(url, headers=headers, json=payload, timeout=120)
print(res.status_code, res.json())
{
    "model": "deepseek-v4-pro",
    "messages": [
        {
            "role": "user",
            "content": "Xin chào, bạn là ai?"
        }
    ]
}

Try it

Send a prompt with your API key — uses real credits, streams back instantly.

Reviews & comments
0.0 / 0 reviews
Loading reviews...
Contact Support
We are always ready to help you