DeepSeek Chat

Call the DeepSeek series models via the OpenAI-compatible format.

Endpoint

POST $URL/v1/chat/completions

Supported Models

Model	Positioning
deepseek-v4-pro	V4 Pro, stronger reasoning
deepseek-v4-flash	V4 high-speed lightweight version

More DeepSeek models will be updated here as they become available.

Request Example

curl

curl $URL/v1/chat/completions \
  -H "Authorization: Bearer $YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "deepseek-v4-flash",
    "messages": [
      {"role": "user", "content": "Explain the Transformer attention mechanism"}
    ]
  }'

Python

from openai import OpenAI

client = OpenAI(api_key="YOUR_API_KEY", base_url="$URL/v1")
resp = client.chat.completions.create(
    model="deepseek-v4-flash",
    messages=[{"role": "user", "content": "Explain the Transformer attention mechanism"}],
)
print(resp.choices[0].message.content)

TypeScript

import OpenAI from 'openai'

const client = new OpenAI({
  apiKey: process.env.YOUR_API_KEY,
  baseURL: '$URL/v1',
})
const resp = await client.chat.completions.create({
  model: 'deepseek-v4-flash',
  messages: [{ role: 'user', content: 'Explain the Transformer attention mechanism' }],
})
console.log(resp.choices[0].message.content)

Notes

DeepSeek models are accessed via the OpenAI-compatible format. The request body, authentication method and response structure are identical to GPT/Claude. Switching models only requires changing the model field.