DeepSeek Language

DeepSeek OCR

OCR model for extracting text, tables from images and PDFs

64K Context window

Text + Vision Modalities

General Specialty

Yes Streaming

Get API key — free View docs →

Model ID deepseek-ocr /v1/chat/completions

Our pricing

Input tokens per 1M tok

$0.046

Output tokens per 1M tok

$0.205

Pay-as-you-go · No minimums · Cancel anytime

OpenAI-compatible ( api.belugapi.com/v1)

Start for free →

No credit card required to test

Capabilities

What DeepSeek OCR
can do for you.

Native API parity with the official provider — every feature surfaced one-to-one.

Streaming

Server-sent events out of the box

Vision

Analyze images, charts & documents

Integration

2-line migration.
Zero friction.

BelugAPI is 100 % compatible with the OpenAI SDK. Point base_url to our endpoint and you're done — no refactoring, no learning curve.

Same request & response schema

Python · Node.js · REST · Go · Ruby

Streaming, tool-calling, structured output

Automatic failover & load balancing

Python

from openai import OpenAI

client = OpenAI(
  api_key=bel-your-key,
  base_url=https://api.belugapi.com/v1
)

response = client.chat.completions.create(
  model=deepseek-ocr,
  messages=[{"role": user, "content": Hello!}]
)
print(response.choices[0].message.content)

Node.js

import OpenAI from openai;

const client = new OpenAI({
  apiKey:  bel-your-key,
  baseURL: https://api.belugapi.com/v1,
});

const res = await client.chat.completions.create({
  model:    deepseek-ocr,
  messages: [{ role: user, content: Hello! }],
});
console.log(res.choices[0].message.content);

cURL

curl https://api.belugapi.com/v1/v1/chat/completions \
  -H Authorization: Bearer bel-your-key \
  -H Content-Type: application/json \
  -d {"model":"deepseek-ocr","messages":[{"role":"user","content":"Hello!"}]}

Highlights

Built for production.

OCR

Document understanding

Complete guide

Everything about
DeepSeek OCR.

Specifications, pricing, capabilities, and integration tips — kept up to date with every DeepSeek release.

By DeepSeek 2 min read Updated May 2026

OCR model for extracting text, tables from images and PDFs

Overview

DeepSeek OCR is a cutting-edge vision ai model developed by DeepSeek, designed to push the boundaries of artificial intelligence-powered content generation.

This model excels in OCR, document understanding, making it a top choice for professionals seeking high-quality, scalable AI solutions. Whether you're building production applications, researching new AI capabilities, or creating stunning visual content, DeepSeek OCR delivers industry-leading performance and reliability.

Key Specifications

Specification	Details
Model Name	DeepSeek OCR
Provider	DeepSeek
Category	Vision AI Model
Model Type	Vision
Context Window	32K

Pricing

Input: $0.0456/M tokens
Output: $0.2052/M tokens

Key Features & Capabilities

Ocr: Advanced capability for professional-grade output.
Document Understanding: Advanced capability for professional-grade output.

Use Cases & Applications

AI-powered automation
Content generation workflows
Enterprise AI solutions
Creative professional tools

Frequently asked questions

What is DeepSeek OCR best used for?

DeepSeek OCR excels at OCR, document understanding, making it ideal for professional and enterprise applications.

Who developed DeepSeek OCR?

DeepSeek OCR was developed by DeepSeek, a leading AI research and development company.

How do I integrate DeepSeek OCR into my application?

You can integrate DeepSeek OCR via its official API endpoint using standard HTTP requests with your API key. SDKs are available for Python, JavaScript, and other languages.

What is the pricing model for DeepSeek OCR?

Pricing is based on input. Check the pricing section above for detailed rates.

Ready to ship? Start using DeepSeek OCR in under 30 seconds.

Get free API key