Layout analysis of many document pages

POST

/v1/batch/documents/parse

curl --request POST \
  --url https://api.stack.foldbun.com/v1/batch/documents/parse \
  --header 'Authorization: Bearer <token>' \
  --header 'Content-Type: application/json' \
  --data '{ "model": "paddleocr-vl", "items": [ { "image": { "id": "img_3b1f0c5e9a7d42c8b6e1f0a9d8c7b6a5f4e3d2c1b0a998877665544332211000" } } ] }'

• Production.

Model paddleocr-vl (default). Items are POST /v1/documents/parse bodies without model. Semantics, modes, limits and the event format are described on the BatchRequest, BatchItem and BatchSummary schemas and in the Overview. Server-Sent Events are recommended; JSON mode is not.

Authorizations

• bearer

Parameters

Header Parameters

Idempotency-Key

string

>= 1 characters <= 128 characters /^[\x21-\x7e]+$/

1 to 128 printable ASCII characters. Within an account, a key is remembered for 24 hours: repeating the request with the same body returns the original response (header Idempotent-Replayed: true) without charging or running the call again. The same key with a different body is 409 idempotency_conflict. See the Overview.

Request Body^required

The model, the items and an optional time limit.

Media typeapplication/json

A batch of single calls to one endpoint. model is set once (omit it where the endpoint has a default model); an item must not contain model. Every item is validated before anything runs: a structural error is 400 invalid_request whose message names items[i], and nothing is billed. Images in items can only be {"id": "img_..."} handles. Items run strictly one after another, so the order of the array is the order of execution. At most 1,000 items and 4 MiB per request. Idempotency-Key makes item i use the derived key "b <key> <i>": repeating the same batch replays finished items (replayed: true, no charge), runs the others, and an item whose content changed fails with idempotency_conflict.

Limits: event-stream mode runs for at most 600 s, JSON mode for at most 90 s (JSON mode is not recommended beyond a few items). timeoutSec can only shorten them; a larger value is clamped without an error. When the limit passes, no new item starts, the running item is cancelled (not billed) and every item not finished is expired; the results of finished items are kept. A client disconnect stops the batch the same way. At most 1 to 4 batch requests can be open per account at once, depending on the tier (429 too_many_batches).

object

model

Model id for all items. Optional where the endpoint has a default model.

string

items

required

Single-call request bodies of the endpoint, without model, in execution order.

Array<object>

>= 1 items <= 1000 items

The request body of the single-call endpoint without model; images only as {"id": "img_..."}.

object

timeoutSec

Time limit for the whole batch in seconds, counted from the arrival of the request. Can only shorten the mode limit (600 s event stream, 90 s JSON).

number

> 0

Example

{
  "model": "paddleocr-vl",
  "items": [
    {
      "image": {
        "id": "img_3b1f0c5e9a7d42c8b6e1f0a9d8c7b6a5f4e3d2c1b0a998877665544332211000"
      }
    }
  ]
}

Responses

200

The batch ran (items may still have failed individually: the request itself only fails before execution, with the normal error envelope). With Accept: text/event-stream the body is a stream of events, flushed one by one: progress first, then item per finished item, summary last. Otherwise it is one JSON document, or one CBOR document with Accept: application/cbor.

Events in the form event: <name> / data: <one-line JSON>, separated by a blank line. Names: progress (BatchProgress), item (BatchItem), summary (BatchSummary, last).

string

Example

event: progress
data: {"done":0,"remaining":2}

event: item
data: {"index":0,"status":"ok","result":{"id":"call_0193...","model":"dinov3","cost":0.05,"balance":9.95}}

event: progress
data: {"done":1,"remaining":1,"etaWaitMs":850}

event: item
data: {"index":1,"status":"error","error":{"code":"insufficient_balance","message":"insufficient balance: this call can cost up to 0.05 credits"}}

event: summary
data: {"counts":{"ok":1,"error":1,"expired":0},"cost":0.05,"balance":0.01,"elapsedMs":2310}

The JSON-mode response (not recommended; see BatchRequest).

object

items

required

One entry per item, sorted by index.

Array<object>

The outcome of one item; the data of an item event and an element of BatchResponse.items.

object

index

required

Position of the item in the request.

integer

status

required

ok: the call succeeded and was billed. error: it failed and was not billed. expired: the batch time limit passed before the item finished; not billed.

string

Allowed values: ok error expired

result

On ok: exactly the success response of the single call, including id, cost, balance, billing and timing.

object

error

On error: the error of the single call (an insufficient balance fails only this item; the next items are still tried).

object

code

required

Stable error code

string

message

required

Human-readable explanation.

string

replayed

True when the item was answered from an earlier batch under the same Idempotency-Key; nothing ran or was charged.

boolean

summary

required

The final summary event, and summary of the JSON response.

object

counts

required

Number of items per status.

object

ok

required

Items with status ok.

integer

error

required

Items with status error.

integer

expired

required

Items with status expired.

integer

cost

required

Credits charged by this request (replayed items are not charged again).

number

balance

required

Account balance when the batch ended.

number

elapsedMs

required

Time from the arrival of the request to the end of the batch in milliseconds.

integer

Example

{
  "items": [
    {
      "status": "ok"
    }
  ]
}

The JSON-mode response (not recommended; see BatchRequest).

object

items

required

One entry per item, sorted by index.

Array<object>

The outcome of one item; the data of an item event and an element of BatchResponse.items.

object

index

required

Position of the item in the request.

integer

status

required

ok: the call succeeded and was billed. error: it failed and was not billed. expired: the batch time limit passed before the item finished; not billed.

string

Allowed values: ok error expired

result

On ok: exactly the success response of the single call, including id, cost, balance, billing and timing.

object

error

On error: the error of the single call (an insufficient balance fails only this item; the next items are still tried).

object

code

required

Stable error code

string

message

required

Human-readable explanation.

string

replayed

True when the item was answered from an earlier batch under the same Idempotency-Key; nothing ran or was charged.

boolean

summary

required

The final summary event, and summary of the JSON response.

object

counts

required

Number of items per status.

object

ok

required

Items with status ok.

integer

error

required

Items with status error.

integer

expired

required

Items with status expired.

integer

cost

required

Credits charged by this request (replayed items are not charged again).

number

balance

required

Account balance when the batch ended.

number

elapsedMs

required

Time from the arrival of the request to the end of the batch in milliseconds.

integer

Headers

X-Request-Id

string

/^[0-9a-f]{16}$/

Server-generated request id (16 hexadecimal characters); present on every response. Quote it when contacting support.

default

The error envelope. The HTTP status follows the code; only retryable: true errors are worth repeating unchanged. A failed call costs 0.

Status	Codes
400	invalid_request, invalid_image (the image passed the upload checks but could not be decoded), unsupported
401	unauthorized
402	insufficient_balance
403	model_locked
404	not_found, image_not_found
409	idempotency_conflict, request_in_progress
413	payload_too_large
429	rate_limited, daily_limit_reached, too_many_batches (all with Retry-After)
500	internal_error
502	upstream_error, bad_model_output
503	model_unavailable, model_busy, unavailable (all with Retry-After)
504	upstream_timeout

model_unavailable also covers a short loss of contact with the only GPU worker serving the model: for about a minute after the loss, interactive calls answer it at once with a short Retry-After (5 s), while sync and batch calls stay queued until their own deadline instead of failing.

Meaning and retry guidance for each code: Errors.

Media typeapplication/json

Body of every error response.

object

error

required

The error.

object

code

required

Stable machine-readable error code. Branch on this, not on message. Listed with the HTTP statuses under Responses.

string

message

required

Human-readable explanation. The wording can change.

string

retryable

required

True when repeating the same request unchanged can succeed.

boolean

retryAfterMs

required

Suggested wait before a retry in milliseconds; 0 when there is no hint.

integer

Example

{
  "error": {
    "code": "model_busy",
    "message": "model sam3 is busy; retry shortly",
    "retryable": true,
    "retryAfterMs": 5000
  }
}

Headers

Retry-After

integer

>= 1

Whole seconds to wait before retrying; present on 429 and 503.

X-Request-Id

string

/^[0-9a-f]{16}$/

Server-generated request id (16 hexadecimal characters); present on every response. Quote it when contacting support.