DeepSeek V4 - AI Model API and Pricing

Where it runs now

DeepSeek V4 runs now at 2 credits per 1k words. Pick it as the model in the AI Article Generator, AI Content Generator or AI Ebook Generator. The exact cost appears before you generate.

Reasoning-heavy writing on a small budget

Its listed strengths are strong reasoning and low token cost. That makes it a fit for explainers, comparison pieces, research summaries and how-to guides. In those, each section must follow from the last. It also suits ebooks that walk a reader through a method step by step.

Prompts that play to this: "Write a 2,000-word guide on choosing a heat pump for an older house. Lay out the decision in order, state assumptions, end with a checklist." Or "Argue for and against a four-day week for a 20-person agency, then give a recommendation."

Softening a technical tone

The listed limit is that the style can read as technical. Tell it who the reader is and ask for plain words, short sentences and one concrete example per section. For consumer-facing pages, a light editing pass helps. Some teams draft the structure here and polish the final copy with a model known for style.

Presets and settings

In the AI Article Generator, SEO article and Blog post match its long-form strength. Newsletter and Video script also work when the piece needs a clear line of reasoning. Ad copy and Product description are short formats where its edge matters less. The engine reads only your prompt, and the tool caps each response at a fixed length. Request very long pieces section by section.

Costs and how it compares

An 800-word post rounds up to 1k words and costs 2 credits, about $0.02 at the best pack. A 2,000-word article costs 4 credits, about $0.04. It ranks No. 2 from the cheapest of 10 text models, which run from 1 to 47 credits per 1k words. For comparison, Gemini 3 Flash costs 5 credits for short descriptions. GLM-5 costs 6 for bilingual copy, and Claude Haiku 4.5 costs 8 for natural headlines and ad variants.

Where to actually run it

DeepSeek V4 API

curl -X POST https://ugesi.com/api/generate \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "deepseek-v4",
    "prompt": "a product shot on white marble"
  }'

<?php
$ch = curl_init('https://ugesi.com/api/generate');
curl_setopt_array($ch, [
    CURLOPT_POST => true,
    CURLOPT_RETURNTRANSFER => true,
    CURLOPT_HTTPHEADER => [
        'Authorization: Bearer YOUR_API_KEY',
        'Content-Type: application/json',
    ],
    CURLOPT_POSTFIELDS => json_encode([
        'model' => 'deepseek-v4',
        'prompt' => 'a product shot on white marble',
    ]),
]);
$response = json_decode(curl_exec($ch), true);
curl_close($ch);
print_r($response);

type GenerateRequest = {
  model: string;
  prompt: string;
};
type GenerateResponse = {
  job_id: number;
  job_token: string;
  credits_held: number;
  status: "queued" | "succeeded";
  output_text?: string;
};

const body: GenerateRequest = {
  model: "deepseek-v4",
  prompt: "a product shot on white marble",
};

const resp = await fetch("https://ugesi.com/api/generate", {
  method: "POST",
  headers: {
    "Authorization": `Bearer ${process.env.UGESI_API_KEY}`,
    "Content-Type": "application/json",
  },
  body: JSON.stringify(body),
});
if (!resp.ok) throw new Error(`HTTP ${resp.status}`);
const job = (await resp.json()) as GenerateResponse;
console.log(job.job_id, job.status);

import requests

resp = requests.post(
    "https://ugesi.com/api/generate",
    headers={"Authorization": "Bearer YOUR_API_KEY"},
    json={
        "model": "deepseek-v4",
        "prompt": "a product shot on white marble",
    },
    timeout=60,
)
resp.raise_for_status()
print(resp.json())

{
  "model": "deepseek-v4",
  "prompt": "a product shot on white marble"
}

The API comes with every credit pack, from Lite to Enterprise, and uses the same balance as the website. API documentation and access →

License and terms of use

DeepSeek V4 is released under the MIT License, a permissive license that allows commercial use. The full rules are in our Terms of service.

People also look for