Document OCR model hosted on our own GPU. Send an image plus a response_format JSON schema to extract fields. Not a conversational model — system prompts and tool calls are not supported.
DeepSeek OCR is a specialized vision model optimized for high-accuracy document extraction within the LLM Resayil ecosystem. Built on a 3.3B parameter architecture with an 8,192 token context window, this model excels at structured data retrieval rather than general conversation. Developers can integrate it immediately by sending an image alongside a response_format JSON schema, ensuring consistent output for automated pipelines without complex prompt engineering. Its F16 quantization balances precision and performance, delivering competitive accuracy on bilingual Arabic and English tasks compared to larger multimodal alternatives.
Designed for production environments, this model operates at a standard 1x credit multiplier, providing cost-effective scalability for high-volume processing tasks. Unlike general-purpose multimodal models, DeepSeek OCR eliminates conversational overhead to focus strictly on field extraction, reducing latency and token consumption. It supports basic tier access, allowing teams to deploy robust document intelligence solutions without enterprise commitments. By handling complex layouts and mixed-language documents efficiently, it serves as a foundational component for applications requiring strict schema adherence and rapid deployment.
from openai import OpenAI
client = OpenAI(
base_url="https://llmapi.resayil.io/v1/",
api_key="YOUR_API_KEY"
)
response = client.chat.completions.create(
model="deepseek-ocr:latest",
messages=[
{"role": "user", "content": "Hello!"}
]
)
print(response.choices[0].message.content)const response = await fetch(
"https://llmapi.resayil.io/v1/chat/completions",
{
method: "POST",
headers: {
"Content-Type": "application/json",
"Authorization": "Bearer YOUR_API_KEY"
},
body: JSON.stringify({
model: "deepseek-ocr:latest",
messages: [
{ role: "user", content: "Hello!" }
]
})
}
);
const data = await response.json();
console.log(data.choices[0].message.content);curl https://llmapi.resayil.io/v1/chat/completions \
-H "Content-Type: application/json" \
-H "Authorization: Bearer YOUR_API_KEY" \
-d '{
"model": "deepseek-ocr:latest",
"messages": [
{"role": "user", "content": "Hello!"}
]
}'