Document Intelligence Overview

Sarvam’s Document Intelligence API provides enterprise-grade document processing powered by Sarvam Vision, our state-of-the-art multimodal model.

Transform any document into structured, searchable, and machine-readable data with world-class accuracy.

What is Document Intelligence?

Document Intelligence is a comprehensive document processing pipeline powered by Sarvam Vision that:

1. Extracts Text: High-fidelity text extraction across 23 languages (22 Indian + English)
2. Preserves Structure: Maintains document layout, reading order, and hierarchies
3. Parses Tables: Transforms tables into structured HTML or Markdown formats
4. Outputs Structured Data: Generates clean, machine-readable HTML or Markdown output

Key Features

Supported Languages

Document Intelligence supports all 22 Constitutionally recognized Indian languages:

Supported Input Formats

Quick Start

Get started with Document Intelligence in minutes:

1
from sarvamai import SarvamAI
2
3
client = SarvamAI(
4
    api_subscription_key="YOUR_SARVAM_API_KEY"
5
)
6
7
# Create a Document Intelligence job
8
job = client.document_intelligence.create_job(
9
    language="hi-IN",           # Target language (BCP-47 format)
10
    output_format="md"          # Output format: "html" or "md" (delivered as ZIP)
11
)
12
13
# Upload your document
14
job.upload_file("document.pdf")
15
16
# Start processing
17
job.start()
18
19
# Wait for completion
20
status = job.wait_until_complete()
21
print(f"Job completed: {status.job_state}")
22
23
# Get processing metrics
24
metrics = job.get_page_metrics()
25
print(f"Pages processed: {metrics['pages_processed']}")
26
27
# Download the output (ZIP file containing the processed document)
28
job.download_output("./output.zip")
29
print("Output saved to ./output.zip")

Response Format

Job Status Response

1
{
2
  "job_id": "abc123-def456-ghi789",
3
  "job_state": "Completed",
4
  "created_at": "2026-02-04T10:30:00Z",
5
  "updated_at": "2026-02-04T10:35:00Z",
6
  "page_metrics": {
7
    "total_pages": 10,
8
    "pages_processed": 10,
9
    "pages_succeeded": 10,
10
    "pages_failed": 0
11
  }
12
}

Job States

Error Handling

1
from sarvamai import SarvamAI
2
from sarvamai.core.api_error import ApiError
3
4
client = SarvamAI(api_subscription_key="YOUR_SARVAM_API_KEY")
5
6
try:
7
    job = client.document_intelligence.create_job(
8
        language="hi-IN",
9
        output_format="md"
10
    )
11
    job.upload_file("document.pdf")
12
    job.start()
13
    status = job.wait_until_complete()
14
    
15
    if status.job_state == "Completed":
16
        job.download_output("./output.zip")
17
        print("Output saved to ./output.zip")
18
    else:
19
        print(f"Job failed: {status}")
20
        
21
except ApiError as e:
22
    if e.status_code == 400:
23
        print(f"Bad request: {e.body}")
24
    elif e.status_code == 403:
25
        print("Invalid API key")
26
    elif e.status_code == 429:
27
        print("Rate limit exceeded")
28
    else:
29
        print(f"Error {e.status_code}: {e.body}")
30
except FileNotFoundError:
31
    print("Document file not found")

Error Codes

Best Practices

Next Steps