> ## Documentation Index > Fetch the complete documentation index at: https://docs.didit.me/llms.txt > Use this file to discover all available pages before exploring further. # Document AI API > Extract the fields **you** define from any supporting document — payslips, proof of funds, source-of-wealth letters, tax statements, contracts, invoices — in one call. Send the file plus a `fields` definition (key, name, extraction instruction, type) and get back typed values (`text`, `number`, `date` as `YYYY-MM-DD`), an optional name-match score against the expected person or company, PDF/EXIF tamper signals, an `Approved`/`Declined`/`In Review` `status`, and a `warnings` list explaining every issue found. This is the server-to-server counterpart of the hosted Document AI workflow step (see `/core-technology/document-ai/overview`); it uses Didit's default extraction model and does not expose provider selection. **Latency.** Extraction is LLM-based: typical calls take **5–15 seconds**, multi-page PDFs up to **~30 seconds**. Configure a client timeout of **at least 45 seconds** and do not retry before the call completes. **Request format.** `multipart/form-data` only. `fields` is **one JSON-encoded array string** (1–30 items, at most 16 KB); bracket-style form arrays are rejected with `400`. Each item is `{key, name, instruction, type, required, is_full_name}` — `key` is the JSON key you get back (`^[a-z0-9_]{1,64}$`, unique), `instruction` tells the model what to read, `type` is `text` (default) | `number` | `date`, `required` raises `DOCUMENT_AI_MISSING_REQUIRED_FIELDS` when the value cannot be read, and at most one field may set `is_full_name: true` to mark the value compared against the expected name. **Name matching.** Pass `expected_first_name` + `expected_last_name` (person) **or** `expected_company_name` (+ optional `expected_company_country`) — not both. The value of the `is_full_name` field (or, when none is marked, the first text field whose key looks like a name) is compared fuzzily; `name_match_score` (0–100) and `name_on_document` are always returned when both sides exist, and a score below `document_ai_name_match_score_threshold` (default 80) adds `DOCUMENT_AI_NAME_MISMATCH`, governed by `document_ai_name_mismatch_action`. **Decision logic.** Each risk maps to a configurable action: `DECLINE` → `status: "Declined"`, `REVIEW` → `status: "In Review"`, `NO_ACTION` → informational only. Defaults mirror the workflow step: unreadable document `REVIEW`, missing required fields `REVIEW`, suspected manipulation `DECLINE`, name mismatch `NO_ACTION`, unsupported file `REVIEW`. A document that cannot be processed at all returns `400`; a readable but problematic document returns `200` with `Declined`/`In Review` — always inspect `document_ai.status` and `document_ai.warnings`, not just the HTTP code. **Billing.** Each `200` response consumes one Document AI API credit — **$0.20 per document, no free tier** (standalone APIs never have one). When the organization's balance cannot cover the call, the endpoint returns `403` before any processing. Sandbox API keys return a deterministic mock response and are not charged. **Persistence.** With `save_api_request=true` (default) the call is stored as an API-type session: the document, the extracted values, the field definitions and the name-match snapshot are kept, the session shows up in the Business Console, and `request_id` works with `GET /v3/session/{sessionId}/decision/`. Send one document and the list of fields you want read from it. Didit runs its default extraction model over the file, returns every field typed (`text`, `number`, or `date` as `YYYY-MM-DD`), checks the file for manipulation, and optionally compares the name on the document against the person or company you expect. Use it for proof of funds, payslips, bank statements, tax documents, invoices, or any document that does not fit a fixed template. This is the server-to-server version of the hosted [Document AI](/core-technology/document-ai/overview) step. The request is `multipart/form-data`: the file goes in `document`, and the field definitions go in `fields` as one JSON-encoded array string. ## Define the fields to extract `fields` is a JSON array of 1 to 30 items, sent as a single string. Each item has a `key` (lowercase letters, digits, and underscores, unique within the request), a human-readable `name`, an `instruction` that tells the model what to look for, a `type` (`text`, `number`, or `date`), and two flags: `required` and `is_full_name`. ```json theme={null} [ { "key": "account_holder", "name": "Account holder", "instruction": "Full name of the person or company the account belongs to", "type": "text", "required": true, "is_full_name": true }, { "key": "balance", "name": "Closing balance", "instruction": "Closing balance shown on the statement, as a number without the currency symbol", "type": "number", "required": true }, { "key": "currency", "name": "Currency", "instruction": "Three-letter currency code of the balance", "type": "text" }, { "key": "statement_date", "name": "Statement date", "instruction": "Date the statement was issued", "type": "date", "required": true } ] ``` In a multipart request the whole array is one form value: ```bash theme={null} curl -X POST https://verification.didit.me/v3/document-ai/ \ -H "x-api-key: $DIDIT_API_KEY" \ -F "document=@./bank-statement.pdf" \ -F 'fields=[{"key":"account_holder","name":"Account holder","instruction":"Full name of the account holder","type":"text","required":true,"is_full_name":true},{"key":"balance","name":"Closing balance","instruction":"Closing balance as a number","type":"number","required":true}]' \ -F "expected_first_name=Jane" \ -F "expected_last_name=Doe" ``` The response echoes the normalized `fields` and returns `extracted_data` keyed by your `key` values, with `null` for anything the model could not find. `name`, `instruction`, and `type` are optional on each item (`type` defaults to `text`); `key` is required. Sending `fields` as bracket-style form keys (`fields[0][key]=...`) or as anything other than a JSON array returns `400` before any extraction runs, so nothing is billed. ## Name matching Mark at most one field with `is_full_name: true` and send who you expect the document to belong to: | You are verifying | Send | The name is compared against | | ----------------- | ------------------------------------------------------------------------------------ | ---------------------------- | | A person | `expected_first_name` + `expected_last_name` | the person's full name | | A company | `expected_company_name` (+ optional `expected_company_country`, a 2-letter ISO code) | the company name | Send one group or the other, not both. When a name is extracted and an expected name is present, the response includes `name_match_score` (0 to 100) and `name_on_document`. A score below `document_ai_name_match_score_threshold` (default `80`) raises the `DOCUMENT_AI_NAME_MISMATCH` warning and applies `document_ai_name_mismatch_action`. If no field is marked `is_full_name`, Didit falls back to the first text field whose key looks like a name (`name`, `holder`, `full_name`, `account_holder`); mark the field explicitly when you want to control which value is compared. When no name value can be found on the document, or no expected name is sent, the name check is skipped and `name_match_score` is `null`. ## Risks and actions Each risk below maps to an action parameter you can set per request: `DECLINE`, `REVIEW`, or `NO_ACTION`. The strictest triggered action decides `document_ai.status` (`Declined` > `In Review` > `Approved`), and every triggered risk is listed in `document_ai.warnings`. The defaults match the hosted workflow step; see [Document AI rules & warnings](/core-technology/document-ai/warnings-document-ai) for how each risk is detected. | Risk | Parameter | Default | | ------------------------------------- | -------------------------------------------- | ----------- | | `DOCUMENT_AI_UNREADABLE_DOCUMENT` | `document_ai_unreadable_document_action` | `REVIEW` | | `DOCUMENT_AI_MISSING_REQUIRED_FIELDS` | `document_ai_missing_required_fields_action` | `REVIEW` | | `DOCUMENT_AI_SUSPECTED_MANIPULATION` | `document_ai_document_tampering_action` | `DECLINE` | | `DOCUMENT_AI_NAME_MISMATCH` | `document_ai_name_mismatch_action` | `NO_ACTION` | | `DOCUMENT_AI_UNSUPPORTED_FILE` | `document_ai_unsupported_file_action` | `REVIEW` | ## Files, saving, and billing * **Accepted files:** PDF (multi-page supported), JPG, JPEG, PNG, TIFF, WebP, or a ZIP containing the document, up to 30 MB. Encrypted PDFs return `400`; upload a decrypted PDF or a photo instead. * **`save_api_request`** (default `true`): the request and its result appear in the Business Console under **Manual Checks**, where your team can review them. Set it to `false` to skip saving. * **`vendor_data`** and **`metadata`** are returned unchanged in the response so you can link the result to your own records. * **Price:** \$0.20 per document, billed when the document is processed and a status is returned. Standalone APIs do not include the free monthly workflow tier. See [Pricing](/getting-started/pricing#standalone-api-pricing-server-to-server). Need the user to upload the document themselves, collect up to 3 documents in one flow, auto-translate the upload screen, or cross-reference extracted fields against ID or questionnaire data? Use the hosted [Document AI workflow step](/core-technology/document-ai/overview) instead. ## OpenAPI ````yaml POST /v3/document-ai/ openapi: 3.0.0 info: version: 3.0.0 title: Didit Verification API description: Identity verification API. Authenticate with x-api-key header. servers: - url: https://verification.didit.me security: [] tags: [] paths: /v3/document-ai/: post: tags: - Standalone APIs summary: Document AI (custom field extraction + tamper checks) description: >- Extract the fields **you** define from any supporting document — payslips, proof of funds, source-of-wealth letters, tax statements, contracts, invoices — in one call. Send the file plus a `fields` definition (key, name, extraction instruction, type) and get back typed values (`text`, `number`, `date` as `YYYY-MM-DD`), an optional name-match score against the expected person or company, PDF/EXIF tamper signals, an `Approved`/`Declined`/`In Review` `status`, and a `warnings` list explaining every issue found. This is the server-to-server counterpart of the hosted Document AI workflow step (see `/core-technology/document-ai/overview`); it uses Didit's default extraction model and does not expose provider selection. **Latency.** Extraction is LLM-based: typical calls take **5–15 seconds**, multi-page PDFs up to **~30 seconds**. Configure a client timeout of **at least 45 seconds** and do not retry before the call completes. **Request format.** `multipart/form-data` only. `fields` is **one JSON-encoded array string** (1–30 items, at most 16 KB); bracket-style form arrays are rejected with `400`. Each item is `{key, name, instruction, type, required, is_full_name}` — `key` is the JSON key you get back (`^[a-z0-9_]{1,64}$`, unique), `instruction` tells the model what to read, `type` is `text` (default) | `number` | `date`, `required` raises `DOCUMENT_AI_MISSING_REQUIRED_FIELDS` when the value cannot be read, and at most one field may set `is_full_name: true` to mark the value compared against the expected name. **Name matching.** Pass `expected_first_name` + `expected_last_name` (person) **or** `expected_company_name` (+ optional `expected_company_country`) — not both. The value of the `is_full_name` field (or, when none is marked, the first text field whose key looks like a name) is compared fuzzily; `name_match_score` (0–100) and `name_on_document` are always returned when both sides exist, and a score below `document_ai_name_match_score_threshold` (default 80) adds `DOCUMENT_AI_NAME_MISMATCH`, governed by `document_ai_name_mismatch_action`. **Decision logic.** Each risk maps to a configurable action: `DECLINE` → `status: "Declined"`, `REVIEW` → `status: "In Review"`, `NO_ACTION` → informational only. Defaults mirror the workflow step: unreadable document `REVIEW`, missing required fields `REVIEW`, suspected manipulation `DECLINE`, name mismatch `NO_ACTION`, unsupported file `REVIEW`. A document that cannot be processed at all returns `400`; a readable but problematic document returns `200` with `Declined`/`In Review` — always inspect `document_ai.status` and `document_ai.warnings`, not just the HTTP code. **Billing.** Each `200` response consumes one Document AI API credit — **$0.20 per document, no free tier** (standalone APIs never have one). When the organization's balance cannot cover the call, the endpoint returns `403` before any processing. Sandbox API keys return a deterministic mock response and are not charged. **Persistence.** With `save_api_request=true` (default) the call is stored as an API-type session: the document, the extracted values, the field definitions and the name-match snapshot are kept, the session shows up in the Business Console, and `request_id` works with `GET /v3/session/{sessionId}/decision/`. operationId: post_v3document-ai requestBody: required: true content: multipart/form-data: schema: type: object required: - document - fields properties: document: type: string format: binary description: >- The document to read. Allowed extensions: `tiff`, `jpg`, `jpeg`, `png`, `pdf`, `webp`, `zip`. Maximum upload size: **30 MB** (larger files are rejected with `400`). Multi-page PDFs are supported (expect higher latency); encrypted PDFs are rejected with `400`. Images are automatically compressed before processing. fields: type: string description: >- One JSON-encoded array string with **1–30** field definitions (maximum 16 KB). Each item: `key` (required, `^[a-z0-9_]{1,64}$`, unique — the key returned in `extracted_data`), `name` (≤100 chars, shown in messages), `instruction` (≤500 chars, what the model should read), `type` (`text` default | `number` | `date`), `required` (boolean, default `false`), `is_full_name` (boolean, default `false`, at most one per request). Bracket-style form arrays, invalid JSON, duplicate keys, or more than one `is_full_name` return `400` before any processing or billing. example: >- [{"key": "employer_name", "name": "Employer name", "instruction": "The employer or company issuing the payslip.", "type": "text", "required": true, "is_full_name": false}, {"key": "employee_name", "name": "Employee name", "instruction": "The full name of the employee as printed on the payslip.", "type": "text", "required": true, "is_full_name": true}, {"key": "net_pay", "name": "Net pay", "instruction": "The net amount paid to the employee for the period.", "type": "number", "required": true, "is_full_name": false}, {"key": "pay_date", "name": "Pay date", "instruction": "The payment date shown on the payslip.", "type": "date", "required": false, "is_full_name": false}] expected_first_name: type: string description: >- First name to verify against the document's name field (fuzzy match, transliteration-aware). Use together with `expected_last_name`; cannot be combined with `expected_company_name`. example: Sophia expected_last_name: type: string description: >- Last name to verify against the document's name field, combined with `expected_first_name` for `name_match_score`. example: Martinez expected_company_name: type: string description: >- Company name to verify against the document's name field (for business documents). Cannot be combined with `expected_first_name` / `expected_last_name` (no example value on purpose — the generated request sample uses the person fields; swap them for this one when verifying a business document). expected_company_country: type: string description: >- Optional ISO 3166-1 alpha-2 country code (two ASCII letters, e.g. `FR`) that tunes company-name normalization for `expected_company_name`. Only used with `expected_company_name`. document_ai_name_match_score_threshold: type: integer minimum: 0 maximum: 100 default: 80 description: >- Minimum `name_match_score` (0–100) to consider the name matched. A lower score adds `DOCUMENT_AI_NAME_MISMATCH`, governed by `document_ai_name_mismatch_action`. example: 80 document_ai_unreadable_document_action: type: string enum: - DECLINE - REVIEW - NO_ACTION default: REVIEW description: >- What to do when no field could be read from the document (`DOCUMENT_AI_UNREADABLE_DOCUMENT`). document_ai_missing_required_fields_action: type: string enum: - DECLINE - REVIEW - NO_ACTION default: REVIEW description: >- What to do when at least one `required` field could not be read (`DOCUMENT_AI_MISSING_REQUIRED_FIELDS`; `additional_data.missing_fields` lists them). document_ai_document_tampering_action: type: string enum: - DECLINE - REVIEW - NO_ACTION default: DECLINE description: >- What to do when PDF/EXIF forensics suspect manipulation — modified after digital signing, known PDF editor, overlay-text manipulation, suspicious re-export, image anomalies (`DOCUMENT_AI_SUSPECTED_MANIPULATION`). document_ai_name_mismatch_action: type: string enum: - DECLINE - REVIEW - NO_ACTION default: NO_ACTION description: >- What to do when `name_match_score` is below the threshold (`DOCUMENT_AI_NAME_MISMATCH`). document_ai_unsupported_file_action: type: string enum: - DECLINE - REVIEW - NO_ACTION default: REVIEW description: >- What to do when the upload is empty or not a processable file (`DOCUMENT_AI_UNSUPPORTED_FILE`). save_api_request: type: boolean default: true description: >- When `true` (default), persists the call as an API-type session — visible in the Business Console, retrievable via `GET /v3/session/{sessionId}/decision/` using the returned `request_id`, with the uploaded document, the extracted values, the field definitions and the name-match snapshot stored. When `false`, nothing is stored and `request_id` is a transient UUID for response correlation only; the call is still billed. example: true vendor_data: type: string description: >- Optional opaque string (your internal user id, email, UUID…) stored on the persisted session and echoed back in the response. Use it to correlate API calls with your own records and to filter sessions later. example: user-123 metadata: type: object additionalProperties: true description: >- Optional JSON object stored with the session (when `save_api_request=true`) and echoed back in the response. In multipart requests, send it as a JSON-encoded string field (e.g. `metadata={"flow":"onboarding"}`) — it is parsed into an object. example: flow: onboarding responses: '200': description: >- Document processed. `document_ai.status` is `Approved`, `Declined`, or `In Review`; every detected issue is itemized in `document_ai.warnings`. A problematic document still returns `200` — inspect the body, not just the HTTP code. When `save_api_request=true`, `request_id` is the persisted session id. content: application/json: examples: Approved: summary: Payslip read cleanly, name matched value: request_id: 9d6a3b1e-2f0c-4a8e-9b7d-1c2e3f4a5b6c document_ai: status: Approved extracted_data: employer_name: Acme Payments Ltd employee_name: Sophia Martinez net_pay: 3120.5 pay_date: '2026-07-31' fields: - key: employer_name name: Employer name instruction: The employer or company issuing the payslip. type: text required: true is_full_name: false - key: employee_name name: Employee name instruction: >- The full name of the employee as printed on the payslip. type: text required: true is_full_name: true - key: net_pay name: Net pay instruction: The net amount paid to the employee for the period. type: number required: true is_full_name: false - key: pay_date name: Pay date instruction: The payment date shown on the payslip. type: date required: false is_full_name: false name_match_score: 100 name_on_document: Sophia Martinez warnings: [] vendor_data: user-123 metadata: flow: onboarding created_at: '2026-08-19T16:40:12.318Z' In Review — missing required field: summary: '`net_pay` could not be read; default action REVIEW' value: request_id: 3f2a9c7b-5d4e-4b1a-8c6f-0e9d8c7b6a5f document_ai: status: In Review extracted_data: employer_name: Acme Payments Ltd employee_name: Sophia Martinez net_pay: null pay_date: '2026-07-31' fields: - key: employer_name name: Employer name instruction: The employer or company issuing the payslip. type: text required: true is_full_name: false - key: employee_name name: Employee name instruction: >- The full name of the employee as printed on the payslip. type: text required: true is_full_name: true - key: net_pay name: Net pay instruction: The net amount paid to the employee for the period. type: number required: true is_full_name: false - key: pay_date name: Pay date instruction: The payment date shown on the payslip. type: date required: false is_full_name: false name_match_score: 100 name_on_document: Sophia Martinez warnings: - risk: DOCUMENT_AI_MISSING_REQUIRED_FIELDS feature: DOCUMENT_AI additional_data: missing_fields: - Net pay log_type: warning short_description: Required fields could not be extracted long_description: >- One or more fields configured as required could not be extracted from the uploaded document. vendor_data: null metadata: null created_at: '2026-08-19T16:41:05.002Z' Declined — suspected manipulation: summary: PDF forensics flagged a known editor; default action DECLINE value: request_id: 7c1d2e3f-4a5b-4c6d-8e9f-0a1b2c3d4e5f document_ai: status: Declined extracted_data: employer_name: Acme Payments Ltd employee_name: Sophia Martinez net_pay: 3120.5 pay_date: '2026-07-31' fields: - key: employer_name name: Employer name instruction: The employer or company issuing the payslip. type: text required: true is_full_name: false - key: employee_name name: Employee name instruction: >- The full name of the employee as printed on the payslip. type: text required: true is_full_name: true - key: net_pay name: Net pay instruction: The net amount paid to the employee for the period. type: number required: true is_full_name: false - key: pay_date name: Pay date instruction: The payment date shown on the payslip. type: date required: false is_full_name: false name_match_score: null name_on_document: Sophia Martinez warnings: - risk: DOCUMENT_AI_SUSPECTED_MANIPULATION feature: DOCUMENT_AI additional_data: detection_method: known_editor processed_by_known_editor: ilovepdf log_type: error short_description: Suspected document manipulation long_description: >- The system detected signs of potential manipulation or editing in the uploaded document. vendor_data: null metadata: null created_at: '2026-08-19T16:42:40.771Z' schema: type: object properties: request_id: type: string format: uuid description: >- Persisted session id when `save_api_request=true` (usable with `GET /v3/session/{sessionId}/decision/`); otherwise a transient correlation UUID. document_ai: type: object properties: status: type: string enum: - Approved - Declined - In Review description: >- `Approved` when no warning resolves to a decline or review; `Declined` when at least one risk maps to `DECLINE`; otherwise `In Review` when at least one maps to `REVIEW`. The `warnings` list explains why. extracted_data: type: object additionalProperties: true description: >- One entry per requested field key. Values are typed per the field `type` — strings for `text`, numbers for `number`, `YYYY-MM-DD` strings for `date` — or `null` when the value could not be read. example: employer_name: Acme Payments Ltd employee_name: Sophia Martinez net_pay: 3120.5 pay_date: '2026-07-31' fields: type: array description: >- The normalized field definitions that were used for this extraction, echoed back (also stored with the session when `save_api_request=true`). items: type: object properties: key: type: string name: type: string instruction: type: string type: type: string enum: - text - number - date required: type: boolean is_full_name: type: boolean name_match_score: type: number nullable: true minimum: 0 maximum: 100 description: >- Fuzzy similarity (0–100) between `name_on_document` and the expected person/company name. Returned whenever both a name value and an expected name exist — on matches **and** mismatches; `null` otherwise. example: 96 name_on_document: type: string nullable: true description: >- The name value taken from the document (the `is_full_name` field, or the first name-like text field) that was compared against the expected name. `null` when no name value was available. example: Sophia Martinez warnings: type: array description: >- Empty on a clean approval. Every entry explains one detected risk; entries with `log_type: "error"` set `status` to `Declined`, entries with `log_type: "warning"` set it to `In Review`. items: type: object properties: risk: type: string enum: - DOCUMENT_AI_UNREADABLE_DOCUMENT - DOCUMENT_AI_MISSING_REQUIRED_FIELDS - DOCUMENT_AI_SUSPECTED_MANIPULATION - DOCUMENT_AI_NAME_MISMATCH - DOCUMENT_AI_UNSUPPORTED_FILE description: >- Machine-readable risk code. Each one is governed by the matching `document_ai_*_action` request option. example: DOCUMENT_AI_MISSING_REQUIRED_FIELDS feature: type: string enum: - DOCUMENT_AI description: >- Feature that raised the warning. Always `DOCUMENT_AI` on this endpoint. additional_data: type: object nullable: true additionalProperties: true description: >- Extra context for the risk: `{"missing_fields": ["Net pay"]}` for missing required fields, `{"name_on_document", "score", "threshold"}` for a name mismatch, and the forensic finding (`detection_method`, producer/editor details) for suspected manipulation. `null` when there is nothing to add. example: missing_fields: - Net pay log_type: type: string enum: - error - warning - information description: >- Severity derived from the configured action: `error` (DECLINE), `warning` (REVIEW), `information` (NO_ACTION). short_description: type: string description: Human-readable one-liner for the risk. example: Required fields could not be extracted long_description: type: string description: >- Human-readable explanation, interpolated with `additional_data` where relevant. vendor_data: type: string nullable: true description: Echo of the `vendor_data` you sent, or `null`. metadata: type: object nullable: true additionalProperties: true description: Echo of the `metadata` object you sent, or `null`. created_at: type: string format: date-time description: When the request was processed (UTC, ISO 8601). '400': description: >- Validation or processing error. Field-level problems return DRF's standard envelope (one array of messages per offending field). Encrypted or unreadable documents and extraction failures return `{"error": ...}`. content: application/json: examples: Missing document: summary: '`document` not included in the form data' value: document: - No file was submitted. Invalid fields payload: summary: '`fields` is not one JSON-encoded array string' value: fields: - fields must be a valid JSON array string. Duplicate field keys: summary: Two items share the same `key` value: fields: - Every fields item must have a unique key. Conflicting expectations: summary: Person and company expectations sent together value: non_field_errors: - >- Provide either person expected-name fields or expected_company_name, not both. Invalid metadata: summary: >- `metadata` is not a JSON object (or contains a NUL character) value: metadata: - Metadata must be a JSON object. Oversized vendor_data: summary: '`vendor_data` over the 2,000-byte cap' value: vendor_data: - vendor_data is too long (maximum 2000 bytes). Encrypted PDF: summary: Password-protected PDF value: error: >- The PDF is encrypted. Please upload a decrypted PDF or a photo instead. Extraction failed: summary: The extraction pipeline could not process the document value: error: - Error extracting Document AI information schema: type: object additionalProperties: true '403': description: >- Permission denied. Returned when the `x-api-key` header is missing, malformed, revoked, or belongs to another environment — and also when the calling organization's balance cannot cover the call. Authentication failures return `403` with `{"detail": ...}`; this API never returns `401`. Credit shortfalls — and contract-billed organizations whose rate card has no Document AI API entry — return `403` with `{"error": ...}` before any document processing happens. content: application/json: examples: Missing or invalid API key: summary: No `x-api-key` header, or the key is invalid/revoked value: detail: You do not have permission to perform this action. Not enough credits: summary: Organization balance cannot cover the call value: error: >- You don't have enough credits to perform this request. Please top up at https://business.didit.me Contract rate missing: summary: >- Contract-billed organization whose rate card has no Document AI API entry value: error: >- Document AI API is not included in your contract rate card. Contact support@didit.me to enable it. schema: type: object properties: detail: type: string error: type: string '429': description: >- Rate limit exceeded. All POST/PATCH/DELETE endpoints share a budget of 300 write requests per minute per API key. The response carries `X-RateLimit-Limit`, `X-RateLimit-Remaining`, `X-RateLimit-Reset`, and `Retry-After` headers. content: application/json: examples: Write rate limit exceeded: summary: More than 300 write requests in the current minute value: detail: >- Write request rate limit exceeded. You can make up to 300 requests per minute. schema: type: object properties: detail: type: string security: - ApiKeyAuth: [] components: securitySchemes: ApiKeyAuth: type: apiKey in: header name: x-api-key ````