Skip to content
Model infoModelsSnapshot: 15 Oct 2026

Contextual AI / Contextual AI

Document Parser for RAG

Managed document-understanding pipeline combining vision, OCR, language models and table-extraction tools.

Document extraction / OCR
availableClosed-source managed componentDocument understandingOCR
Official documentation (opens in a new tab)Reviewed 15 Oct 2026
10 of 21 specification fields collected
Context windowNot collectedAwaiting source review
Maximum outputNot collectedAwaiting source review
Input billingSee ratesContextual AI API
Output billingSee ratesPublished units and conditions below

Specifications

Provider-documented facts. “Not collected” means we have not verified that field yet, not that the feature is unsupported.

Context / input limit
Not collected

Exact version-specific contract not established. Unknown does not mean unsupported.

Maximum input tokens
Not collected

Exact version-specific contract not established. Unknown does not mean unsupported.

Maximum output
Not collected

Exact version-specific contract not established. Unknown does not mean unsupported.

Parameters
Not collected

Exact version-specific contract not established. Unknown does not mean unsupported.

Release date
2025-05-13Source for Release date (opens in a new tab)

Public component API general availability; not a single underlying model release date.

Knowledge cutoff
Not collected

Exact version-specific contract not established. Unknown does not mean unsupported.

Reasoning
Not collected

Exact version-specific contract not established. Unknown does not mean unsupported.

Tool calling
Not collected

Exact version-specific contract not established. Unknown does not mean unsupported.

Structured output
SupportedSource for Structured output (opens in a new tab)

Schema for extracted document blocks and metadata, not general JSON generation.

Streaming
Not collected

Exact version-specific contract not established. Unknown does not mean unsupported.

Batch processing
Not collected

Exact version-specific contract not established. Unknown does not mean unsupported.

Fine-tuning
Not collected

Exact version-specific contract not established. Unknown does not mean unsupported.

Context caching
Not collected

Exact version-specific contract not established. Unknown does not mean unsupported.

Input modalities
Documents, ImagesSource for Input modalities (opens in a new tab)
Output modalities
Markdown, Structured document blocksSource for Output modalities (opens in a new tab)
Document input formats
PDF, DOC, DOCX, PPT, PPTX, PNG, JPG, JPEGSource for Document input formats (opens in a new tab)
Document file limit
Less than 300 MBSource for Document file limit (opens in a new tab)
Document page limits
Less than 2,000 pages; Office-to-PDF conversion can change page count.Source for Document page limits (opens in a new tab)
Output formats
markdown-document, markdown-per-page, blocks-per-pageSource for Output formats (opens in a new tab)
Document extraction features
Basic mode handles text-only documents; Standard handles scanned/complex documents and images. Document hierarchy is beta and requires Standard plus a continuous page range starting at page zero. Table splitting is also unavailable in Basic.Source for Document extraction features (opens in a new tab)
Architecture
Orchestrated custom vision, OCR, vision-language models and specialized table-extraction tools.Source for Architecture (opens in a new tab)

Access & pricing

Prices belong to an endpoint and billing tier. A provider listing does not guarantee access for every account or region.

Contextual AI API

Provider source (opens in a new tab)
API model ID
Not collected

This managed task has no independently confirmed fixed model selector.

Access
Upload raw_file with optional processing settings. Returns a job ID for polling and result retrieval.
Regions
Public API serving region not established. Enterprise VPC deployment is offered under separate terms.
API endpoint
POST https://api.contextual.ai/v1/parseSource for API endpoint (opens in a new tab)

Prices retain their published currencies and billing units below.

Published billing units · original currency · observed 15 Oct 2026
CategoryUsageRateUnit
Document parsingBasic pages1$3.001,000 pages
Document parsingStandard pages2$40.001,000 pages

Rate conditions

  1. Published Parse component price for Basic mode; not a text-token or per-question tariff. Page count may change after conversion.
  2. Published Parse component price for Standard mode; not a text-token or per-question tariff. Page count may change after conversion.
Pricing source (opens in a new tab)

Data & enterprise terms

Service terms permit processing Customer Data to deliver and improve the service during and after the term. This is not a universal no-training or zero-retention commitment. Personal-data retention is purpose-dependent; enterprise/VPC agreements and task-result retention require separate review.

Read the service policy (opens in a new tab)

Contextual AI API

Provider source (opens in a new tab)
API model ID
Not collected

This managed task has no independently confirmed fixed model selector.

Access
Retrieve previously created job output in one or more documented formats; not independent inference.
Regions
Public API serving region not established. Enterprise VPC deployment is offered under separate terms.
API endpoint
GET https://api.contextual.ai/v1/parse/jobs/{job_id}/resultsSource for API endpoint (opens in a new tab)

Endpoint pricing has not been collected. Check the provider for applicable rates.

Data & enterprise terms

Service terms permit processing Customer Data to deliver and improve the service during and after the term. This is not a universal no-training or zero-retention commitment. Personal-data retention is purpose-dependent; enterprise/VPC agreements and task-result retention require separate review.

Read the service policy (opens in a new tab)

Caching

Caching can reuse earlier input processing. Availability, lifetime and billing depend on the endpoint. A cache hit still generates a new response.

01 / INITIAL INPUTProcess & retain

The system processes eligible input and retains reusable state.

02 / LATER REQUESTCheck available state

Reuse depends on matching input, session state and the cache policy.

03 / CACHE HITReuse & generate

Reusing state can reduce repeated work. Any charges follow the billing rules below.

Contextual AI API

Not verified

No exact model-specific prompt-cache mechanism, eligibility or discounted tariff established.

How it works
Not collected

No exact model-specific prompt-cache mechanism, eligibility or discounted tariff established.

Minimum reusable input
Not collected

No exact model-specific prompt-cache mechanism, eligibility or discounted tariff established.

Lifetime & refresh
Not collected

No exact model-specific prompt-cache mechanism, eligibility or discounted tariff established.

How to enable it
Not collected

No exact model-specific prompt-cache mechanism, eligibility or discounted tariff established.

Usage fields
Not collected

No exact model-specific prompt-cache mechanism, eligibility or discounted tariff established.

Billing rules
Not collected

No exact model-specific prompt-cache mechanism, eligibility or discounted tariff established.

Limits & data boundaries
Not collected

No exact model-specific prompt-cache mechanism, eligibility or discounted tariff established.

No verified cache rate for this endpoint. An empty rate is not a zero charge.

Caching reviewed 15 Oct 2026

Contextual AI API

Not verified

No exact model-specific prompt-cache mechanism, eligibility or discounted tariff established.

How it works
Not collected

No exact model-specific prompt-cache mechanism, eligibility or discounted tariff established.

Minimum reusable input
Not collected

No exact model-specific prompt-cache mechanism, eligibility or discounted tariff established.

Lifetime & refresh
Not collected

No exact model-specific prompt-cache mechanism, eligibility or discounted tariff established.

How to enable it
Not collected

No exact model-specific prompt-cache mechanism, eligibility or discounted tariff established.

Usage fields
Not collected

No exact model-specific prompt-cache mechanism, eligibility or discounted tariff established.

Billing rules
Not collected

No exact model-specific prompt-cache mechanism, eligibility or discounted tariff established.

Limits & data boundaries
Not collected

No exact model-specific prompt-cache mechanism, eligibility or discounted tariff established.

No verified cache rate for this endpoint. An empty rate is not a zero charge.

Caching reviewed 15 Oct 2026

What the measurements say

No benchmark observations have been added for this model. We do not turn a vendor description into a performance score.

Latency and throughput depend on the host, request size, reasoning settings and load. There is no measured speed or global usage figure in this record.

The record behind the model

Why it is included

Managed model-backed component with service access under retained provider rights. No distinct independently trained checkpoint is asserted.

Classification evidence (opens in a new tab)

Sources reviewed