DeepSeek

DeepSeek V4 Pro

The flagship DeepSeek V4: reasoning and code

Text

Reasoning

Tools

Type: Chat
Context: 1.0M tokens
Russian: high
Input price: 37,63 ₽ / 1M tokens
Output price: 75,25 ₽ / 1M tokens
Release date: April 2026

chat/completions

import OpenAI from "openai"
const client = new OpenAI({
  baseURL: "https://api.yenisei.ru/v1",
  apiKey: process.env.YENISEI_API_KEY,
})
const stream = await client.chat.completions.create({
  model: "deepseek-pro",
  messages: [{ role: "user", content: "Hi!" }],
  stream: true,
})
for await (const chunk of stream) {
  process.stdout.write(chunk.choices[0]?.delta?.content ?? "")
}

About the model

DeepSeek V4 Pro is the flagship model of the lineup for complex tasks that call for detailed analysis. It lays out step-by-step reasoning, works through multi-stage conditions and follows long, detailed instructions. It is especially strong with code: architecture design, debugging, review and explaining complex algorithms. It supports adjustable reasoning depth (reasoning_effort), so the same model can be tuned for a quick answer or a thorough analysis. It calls tools reliably (function-calling) and is suited to agents and automation. Choose it when reasoning quality matters more than speed and price.