Overview
Upstage document APIs let AI agents read documents on demand. An agent sends a document, receives a price quote based on the page count, pays it through AWS Marketplace (x402), and gets the result in the same call flow. Charges appear on the buyer's AWS bill.
Document Parse
Transform unstructured PDFs and scanned files into clean, layout-preserving HTML or Markdown - ready for LLM pipelines, RAG indexing, summarization, or search.
- LLM-Ready Output: Preserves tables, charts, layout, and visual context.
- Layout-Aware Parsing: Handles rotated pages, checkbox states, multi-page tables, and embedded visuals.
- High Accuracy: Industry-leading performance on DP-Bench (TEDS score 93+).
- Up to 100 pages per request.
Information Extract
Extract key data from documents as structured JSON aligned to your schema.
- Schema-Agnostic Flexibility: Adapts to any schema without customization.
- Hidden and Implied Data Extraction: Surfaces visible text, inferred values, and line-item calculations.
- No Fine-Tuning Required: Works out of the box without templates or model adjustments.
- Up to 30 pages per request.
Document Classify
Classify documents into the categories you define.
Pricing
The price is calculated per page before processing and returned in the x402 quote, so the agent knows the exact cost before it pays.
- Document Parse: $0.006 per page (standard), $0.036 per page (enhanced)
- Information Extract: $0.048 per page (standard), $0.072 per page (enhanced)
- Document Classify: $0.0048 per page
PDF and image files are supported.
Highlights
- Pay-per-request document AI for agents - No subscription, sign-up or API key. The agent pays the quoted price over x402 and the charge appears on the AWS bill.
- Price known before processing - The quote is page count x unit price, calculated before the document is processed.
- Production Upstage document models - Layout-preserving Document Parse, schema-aligned Information Extract and Document Classify through one endpoint.
Details
Features and programs
Financing for AWS Marketplace purchases
Pricing
Dimension | Description | Cost/unit |
|---|---|---|
Agentic Commerce x402 | Agentic Commerce meters this unit to bill buyers the seller's x402 quote price. | $1.00 |
Vendor refund policy
We currently do not support refunds, but you can ask through the support channel (https://get.support.upstage.ai/servicedesk/customer/portals ).
How can we make this page better?
Legal
Vendor terms and conditions
Content disclaimer
Delivery details
API-Based Agents & Tools
API-Based Agents and Tools integrate through standard web protocols. Your applications can make API calls to access agent capabilities and receive responses.
Additional details
Usage instructions
API
How to pay for this API
This API is sold pay-per-request through AWS Marketplace using x402. Charges are billed in USD to your AWS account under AWS Marketplace. To call it, set up Amazon Bedrock AgentCore payments in your AWS account (a Payment Manager and a payment session with a spend limit), which your agent uses to obtain the AWS Marketplace payment token.
Send your request to the endpoint below. The API returns HTTP 402 Payment Required with an AWS Marketplace x402 price quote (extra.quoteToken) in the PAYMENT-REQUIRED header. Pass that 402 response to AgentCore payments (ProcessPayment, or the AgentCore payments plugin for Strands/LangGraph), then retry the same request with the payment token in the PAYMENT-SIGNATURE header. A successful call returns 200 with your result, and the PAYMENT-RESPONSE header contains your AWS Marketplace payment ID.
How to call this API
Endpoint: <https://x402.api.upstage.ai>
| Method + URL | API | Request body | Max pages per request |
|---|---|---|---|
| POST <https://x402.api.upstage.ai/v1/document-digitization> | Document Parse | multipart/form-data | 100 |
| POST <https://x402.api.upstage.ai/v1/information-extraction> | Information Extract | application/json | 30 |
| POST <https://x402.api.upstage.ai/v1/document-classification> | Document Classify | application/json | 100 |
No API key or Authorization header is needed. The first request returns 402 Payment Required; pay as described above and send the same request again with the PAYMENT-SIGNATURE header. Send the same document on the retry. The price is the page count times the unit price, so a different document changes the price and the payment is refused. PDF and image files are supported.
Document Parse
Parameters: document (file, required), model (document-parse), mode (standard or enhanced), ocr (auto or force), output_formats (e.g. ["html", "markdown"]). Parameter details: Document Parse API .
curl -X POST <https://x402.api.upstage.ai/v1/document-digitization> \ -F "document=@./invoice.pdf" \ -F "model=document-parse" \ -F "mode=standard" \ -F "output_formats=[\"html\", \"markdown\"]"Information Extract
The body is the same as the Information Extract API : model (information-extract), messages with the document as a base64 data URL, and response_format with your JSON schema.
curl -X POST <https://x402.api.upstage.ai/v1/information-extraction> \ -H "Content-Type: application/json" \ -d '{ "model": "information-extract", "messages": [{"role": "user", "content": [{"type": "image_url", "image_url": {"url": "data:application/octet-stream;base64,<BASE64_DOCUMENT>"}}]}], "response_format": {"type": "json_schema", "json_schema": {"name": "document_schema", "schema": {"type": "object", "properties": { "bank_name": {"type": "string", "description": "The name of the bank"}}}}} }'Document Classify
The body is the same as the Document Classify API , with model set to document-classify.
Errors returned before payment
- 400 / 404: missing or invalid parameters, or an unknown path. No quote is issued.
- 413: the document has more pages than the per-request limit.
- 415: unsupported file type, or the page count cannot be read.
Support
Vendor support
If you encounter any issues or have questions while using it, please contact us
AWS infrastructure support
AWS Support is a one-on-one, fast-response support channel that is staffed 24x7x365 with experienced and technical support engineers. The service helps customers of all sizes and technical abilities to successfully utilize the products and features provided by Amazon Web Services.
