Skip to content
LlamaIndex OSS Documentation
API Reference
Parsing

Parse File

parsing.create(ParsingCreateParams**kwargs) -> ParsingCreateResponse
POST/api/v2/parse

Parse a file by file ID or URL.

Provide either file_id (a previously uploaded file) or source_url (a publicly accessible URL). Configure parsing with options like tier, target_pages, and lang.

Tiers

  • fast — rule-based, cheapest, no AI
  • cost_effective — balanced speed and quality
  • agentic — full AI-powered parsing
  • agentic_plus — premium AI with specialized features

The job runs asynchronously. Poll GET /parse/{job_id} with expand=text or expand=markdown to retrieve results.

ParametersExpand Collapse
tier: Literal["fast", "cost_effective", "agentic", "agentic_plus"]

Parsing tier: 'fast' (rule-based, cheapest), 'cost_effective' (balanced), 'agentic' (AI-powered with custom prompts), or 'agentic_plus' (premium AI with highest accuracy)

Accepts one of the following:
"fast"
"cost_effective"
"agentic"
"agentic_plus"
version: Union[Literal["2025-12-11", "2025-12-18", "2025-12-31", 31 more], str]

Tier version. Use 'latest' for the current stable version, or specify a specific version (e.g., '1.0', '2.0') for reproducible results

Accepts one of the following:
Literal["2025-12-11", "2025-12-18", "2025-12-31", 31 more]

Tier version. Use 'latest' for the current stable version, or specify a specific version (e.g., '1.0', '2.0') for reproducible results

Accepts one of the following:
"2025-12-11"
"2025-12-18"
"2025-12-31"
"2026-01-08"
"2026-01-09"
"2026-01-16"
"2026-01-21"
"2026-01-22"
"2026-01-24"
"2026-01-29"
"2026-01-30"
"2026-02-03"
"2026-02-18"
"2026-02-20"
"2026-02-24"
"2026-02-26"
"2026-03-02"
"2026-03-03"
"2026-03-04"
"2026-03-05"
"2026-03-09"
"2026-03-10"
"2026-03-11"
"2026-03-12"
"2026-03-17"
"2026-03-19"
"2026-03-20"
"2026-03-22"
"2026-03-23"
"2026-03-24"
"2026-03-25"
"2026-03-26"
"2026-03-27"
"latest"
str
organization_id: Optional[str]
project_id: Optional[str]
agentic_options: Optional[AgenticOptions]

Options for AI-powered parsing tiers (cost_effective, agentic, agentic_plus).

These options customize how the AI processes and interprets document content. Only applicable when using non-fast tiers.

custom_prompt: Optional[str]

Custom instructions for the AI parser. Use to guide extraction behavior, specify output formatting, or provide domain-specific context. Example: 'Extract financial tables with currency symbols. Format dates as YYYY-MM-DD.'

client_name: Optional[str]

Identifier for the client/application making the request. Used for analytics and debugging. Example: 'my-app-v2'

crop_box: Optional[CropBox]

Crop boundaries to process only a portion of each page. Values are ratios 0-1 from page edges

bottom: Optional[float]

Bottom boundary as ratio (0-1). 0=top edge, 1=bottom edge. Content below this line is excluded

maximum1
minimum0
left: Optional[float]

Left boundary as ratio (0-1). 0=left edge, 1=right edge. Content left of this line is excluded

maximum1
minimum0
right: Optional[float]

Right boundary as ratio (0-1). 0=left edge, 1=right edge. Content right of this line is excluded

maximum1
minimum0
top: Optional[float]

Top boundary as ratio (0-1). 0=top edge, 1=bottom edge. Content above this line is excluded

maximum1
minimum0
disable_cache: Optional[bool]

Bypass result caching and force re-parsing. Use when document content may have changed or you need fresh results

fast_options: Optional[object]

Options for fast tier parsing (rule-based, no AI).

Fast tier uses deterministic algorithms for text extraction without AI enhancement. It's the fastest and most cost-effective option, best suited for simple documents with standard layouts. Currently has no configurable options but reserved for future expansion.

file_id: Optional[str]

ID of an existing file in the project to parse. Mutually exclusive with source_url

http_proxy: Optional[str]

HTTP/HTTPS proxy for fetching source_url. Ignored if using file_id

input_options: Optional[InputOptions]

Format-specific options (HTML, PDF, spreadsheet, presentation). Applied based on detected input file type

html: Optional[InputOptionsHTML]

HTML/web page parsing options (applies to .html, .htm files)

make_all_elements_visible: Optional[bool]

Force all HTML elements to be visible by overriding CSS display/visibility properties. Useful for parsing pages with hidden content or collapsed sections

remove_fixed_elements: Optional[bool]

Remove fixed-position elements (headers, footers, floating buttons) that appear on every page render

remove_navigation_elements: Optional[bool]

Remove navigation elements (nav bars, sidebars, menus) to focus on main content

pdf: Optional[object]

PDF-specific parsing options (applies to .pdf files)

presentation: Optional[InputOptionsPresentation]

Presentation parsing options (applies to .pptx, .ppt, .odp, .key files)

out_of_bounds_content: Optional[bool]

Extract content positioned outside the visible slide area. Some presentations have hidden notes or content that extends beyond slide boundaries

skip_embedded_data: Optional[bool]

Skip extraction of embedded chart data tables. When true, only the visual representation of charts is captured, not the underlying data

spreadsheet: Optional[InputOptionsSpreadsheet]

Spreadsheet parsing options (applies to .xlsx, .xls, .csv, .ods files)

detect_sub_tables_in_sheets: Optional[bool]

Detect and extract multiple tables within a single sheet. Useful when spreadsheets contain several data regions separated by blank rows/columns

force_formula_computation_in_sheets: Optional[bool]

Compute formula results instead of extracting formula text. Use when you need calculated values rather than formula definitions

include_hidden_sheets: Optional[bool]

Parse hidden sheets in addition to visible ones. By default, hidden sheets are skipped

output_options: Optional[OutputOptions]

Output formatting options for markdown, text, and extracted images

extract_printed_page_number: Optional[bool]

Extract the printed page number as it appears in the document (e.g., 'Page 5 of 10', 'v', 'A-3'). Useful for referencing original page numbers

images_to_save: Optional[List[Literal["screenshot", "embedded", "layout"]]]

Image categories to extract and save. Options: 'screenshot' (full page renders useful for visual QA), 'embedded' (images found within the document), 'layout' (cropped regions from layout detection like figures and diagrams). Empty list saves no images

Accepts one of the following:
"screenshot"
"embedded"
"layout"
markdown: Optional[OutputOptionsMarkdown]

Markdown formatting options including table styles and link annotations

Add link annotations to markdown output in the format text. When false, only the link text is included

inline_images: Optional[bool]

Embed images directly in markdown as base64 data URIs instead of extracting them as separate files. Useful for self-contained markdown output

tables: Optional[OutputOptionsMarkdownTables]

Table formatting options including markdown vs HTML format and merging behavior

compact_markdown_tables: Optional[bool]

Remove extra whitespace padding in markdown table cells for more compact output

markdown_table_multiline_separator: Optional[str]

Separator string for multiline cell content in markdown tables. Example: '
' to preserve line breaks, ' ' to join with spaces

merge_continued_tables: Optional[bool]

Automatically merge tables that span multiple pages into a single table. The merged table appears on the first page with merged_from_pages metadata

output_tables_as_markdown: Optional[bool]

Output tables as markdown pipe tables instead of HTML

tags. Markdown tables are simpler but cannot represent complex structures like merged cells

spatial_text: Optional[OutputOptionsSpatialText]

Spatial text output options for preserving document layout structure

do_not_unroll_columns: Optional[bool]

Keep multi-column layouts intact instead of linearizing columns into sequential text. Automatically enabled for non-fast tiers

preserve_layout_alignment_across_pages: Optional[bool]

Maintain consistent text column alignment across page boundaries. Automatically enabled for document-level parsing modes

preserve_very_small_text: Optional[bool]

Include text below the normal size threshold. Useful for footnotes, watermarks, or fine print that might otherwise be filtered out

tables_as_spreadsheet: Optional[OutputOptionsTablesAsSpreadsheet]

Options for exporting tables as XLSX spreadsheets

enable: Optional[bool]

Whether this option is enabled

guess_sheet_name: Optional[bool]

Automatically generate descriptive sheet names from table context (headers, surrounding text) instead of using generic names like 'Table_1'

page_ranges: Optional[PageRanges]

Page selection: limit total pages or specify exact pages to process

max_pages: Optional[int]

Maximum number of pages to process. Pages are processed in order starting from page 1. If both max_pages and target_pages are set, target_pages takes precedence

minimum1
target_pages: Optional[str]

Comma-separated list of specific pages to process using 1-based indexing. Supports individual pages and ranges. Examples: '1,3,5' (pages 1, 3, 5), '1-5' (pages 1 through 5 inclusive), '1,3,5-8,10' (pages 1, 3, 5-8, and 10). Pages are sorted and deduplicated automatically. Duplicate pages cause an error

processing_control: Optional[ProcessingControl]

Job execution controls including timeouts and failure thresholds

job_failure_conditions: Optional[ProcessingControlJobFailureConditions]

Quality thresholds that determine when a job should fail vs complete with partial results

allowed_page_failure_ratio: Optional[float]

Maximum ratio of pages allowed to fail before the job fails (0-1). Example: 0.1 means job fails if more than 10% of pages fail. Default is 0.05 (5%)

maximum1
exclusiveMinimum0
fail_on_buggy_font: Optional[bool]

Fail the job if a problematic font is detected that may cause incorrect text extraction. Buggy fonts can produce garbled or missing characters

fail_on_image_extraction_error: Optional[bool]

Fail the entire job if any embedded image cannot be extracted. By default, image extraction errors are logged but don't fail the job

fail_on_image_ocr_error: Optional[bool]

Fail the entire job if OCR fails on any image. By default, OCR errors result in empty text for that image

fail_on_markdown_reconstruction_error: Optional[bool]

Fail the entire job if markdown cannot be reconstructed for any page. By default, failed pages use fallback text extraction

timeouts: Optional[ProcessingControlTimeouts]

Timeout settings for job execution. Increase for large or complex documents

base_in_seconds: Optional[int]

Base timeout for the job in seconds (max 1800 = 30 minutes). This is the minimum time allowed regardless of document size

maximum1800
exclusiveMinimum0
extra_time_per_page_in_seconds: Optional[int]

Additional timeout per page in seconds (max 300 = 5 minutes). Total timeout = base + (this value × page count)

maximum300
exclusiveMinimum0
processing_options: Optional[ProcessingOptions]

Document processing options including OCR, table extraction, and chart parsing

aggressive_table_extraction: Optional[bool]

Use aggressive heuristics to detect table boundaries, even without visible borders. Useful for documents with borderless or complex tables

auto_mode_configuration: Optional[Iterable[ProcessingOptionsAutoModeConfiguration]]

Conditional processing rules that apply different parsing options based on page content, document structure, or filename patterns. Each entry defines trigger conditions and the parsing configuration to apply when triggered

parsing_conf: ProcessingOptionsAutoModeConfigurationParsingConf

Parsing configuration to apply when trigger conditions are met

adaptive_long_table: Optional[bool]

Whether to use adaptive long table handling

aggressive_table_extraction: Optional[bool]

Whether to use aggressive table extraction

crop_box: Optional[ProcessingOptionsAutoModeConfigurationParsingConfCropBox]

Crop box options for auto mode parsing configuration.

bottom: Optional[float]

Bottom boundary of crop box as ratio (0-1)

maximum1
minimum0
left: Optional[float]

Left boundary of crop box as ratio (0-1)

maximum1
minimum0
right: Optional[float]

Right boundary of crop box as ratio (0-1)

maximum1
minimum0
top: Optional[float]

Top boundary of crop box as ratio (0-1)

maximum1
minimum0
custom_prompt: Optional[str]

Custom AI instructions for matched pages. Overrides the base custom_prompt

extract_layout: Optional[bool]

Whether to extract layout information

high_res_ocr: Optional[bool]

Whether to use high resolution OCR

ignore: Optional[ProcessingOptionsAutoModeConfigurationParsingConfIgnore]

Ignore options for auto mode parsing configuration.

ignore_diagonal_text: Optional[bool]

Whether to ignore diagonal text in the document

ignore_hidden_text: Optional[bool]

Whether to ignore hidden text in the document

language: Optional[str]

Primary language of the document

outlined_table_extraction: Optional[bool]

Whether to use outlined table extraction

presentation: Optional[ProcessingOptionsAutoModeConfigurationParsingConfPresentation]

Presentation-specific options for auto mode parsing configuration.

out_of_bounds_content: Optional[bool]

Extract out of bounds content in presentation slides

skip_embedded_data: Optional[bool]

Skip extraction of embedded data for charts in presentation slides

spatial_text: Optional[ProcessingOptionsAutoModeConfigurationParsingConfSpatialText]

Spatial text options for auto mode parsing configuration.

do_not_unroll_columns: Optional[bool]

Keep column structure intact without unrolling

preserve_layout_alignment_across_pages: Optional[bool]

Preserve text alignment across page boundaries

preserve_very_small_text: Optional[bool]

Include very small text in spatial output

specialized_chart_parsing: Optional[Literal["agentic_plus", "agentic", "efficient"]]

Enable specialized chart parsing with the specified mode

Accepts one of the following:
"agentic_plus"
"agentic"
"efficient"
tier: Optional[Literal["fast", "cost_effective", "agentic", "agentic_plus"]]

Override the parsing tier for matched pages. Must be paired with version

Accepts one of the following:
"fast"
"cost_effective"
"agentic"
"agentic_plus"
version: Optional[Union[Literal["2025-12-11", "2025-12-18", "2025-12-31", 31 more], str, null]]

Tier version when overriding tier. Required when tier is specified

Accepts one of the following:
Literal["2025-12-11", "2025-12-18", "2025-12-31", 31 more]

Tier version when overriding tier. Required when tier is specified

Accepts one of the following:
"2025-12-11"
"2025-12-18"
"2025-12-31"
"2026-01-08"
"2026-01-09"
"2026-01-16"
"2026-01-21"
"2026-01-22"
"2026-01-24"
"2026-01-29"
"2026-01-30"
"2026-02-03"
"2026-02-18"
"2026-02-20"
"2026-02-24"
"2026-02-26"
"2026-03-02"
"2026-03-03"
"2026-03-04"
"2026-03-05"
"2026-03-09"
"2026-03-10"
"2026-03-11"
"2026-03-12"
"2026-03-17"
"2026-03-19"
"2026-03-20"
"2026-03-22"
"2026-03-23"
"2026-03-24"
"2026-03-25"
"2026-03-26"
"2026-03-27"
"latest"
str
filename_match_glob: Optional[str]

Single glob pattern to match against filename

filename_match_glob_list: Optional[SequenceNotStr[str]]

List of glob patterns to match against filename

filename_regexp: Optional[str]

Regex pattern to match against filename

filename_regexp_mode: Optional[str]

Regex mode flags (e.g., 'i' for case-insensitive)

full_page_image_in_page: Optional[bool]

Trigger if page contains a full-page image (scanned page detection)

full_page_image_in_page_threshold: Optional[Union[float, str, null]]

Threshold for full page image detection (0.0-1.0, default 0.8)

Accepts one of the following:
float
str
image_in_page: Optional[bool]

Trigger if page contains non-screenshot images

layout_element_in_page: Optional[str]

Trigger if page contains this layout element type

layout_element_in_page_confidence_threshold: Optional[Union[float, str, null]]

Confidence threshold for layout element detection

Accepts one of the following:
float
str
page_contains_at_least_n_charts: Optional[Union[int, str, null]]

Trigger if page has more than N charts

Accepts one of the following:
int
str
page_contains_at_least_n_images: Optional[Union[int, str, null]]

Trigger if page has more than N images

Accepts one of the following:
int
str
page_contains_at_least_n_layout_elements: Optional[Union[int, str, null]]

Trigger if page has more than N layout elements

Accepts one of the following:
int
str
page_contains_at_least_n_lines: Optional[Union[int, str, null]]

Trigger if page has more than N lines

Accepts one of the following:
int
str
page_contains_at_least_n_links: Optional[Union[int, str, null]]

Trigger if page has more than N links

Accepts one of the following:
int
str
page_contains_at_least_n_numbers: Optional[Union[int, str, null]]

Trigger if page has more than N numeric words

Accepts one of the following:
int
str
page_contains_at_least_n_percent_numbers: Optional[Union[int, str, null]]

Trigger if page has more than N% numeric words

Accepts one of the following:
int
str
page_contains_at_least_n_tables: Optional[Union[int, str, null]]

Trigger if page has more than N tables

Accepts one of the following:
int
str
page_contains_at_least_n_words: Optional[Union[int, str, null]]

Trigger if page has more than N words

Accepts one of the following:
int
str
page_contains_at_most_n_charts: Optional[Union[int, str, null]]

Trigger if page has fewer than N charts

Accepts one of the following:
int
str
page_contains_at_most_n_images: Optional[Union[int, str, null]]

Trigger if page has fewer than N images

Accepts one of the following:
int
str
page_contains_at_most_n_layout_elements: Optional[Union[int, str, null]]

Trigger if page has fewer than N layout elements

Accepts one of the following:
int
str
page_contains_at_most_n_lines: Optional[Union[int, str, null]]

Trigger if page has fewer than N lines

Accepts one of the following:
int
str
page_contains_at_most_n_links: Optional[Union[int, str, null]]

Trigger if page has fewer than N links

Accepts one of the following:
int
str
page_contains_at_most_n_numbers: Optional[Union[int, str, null]]

Trigger if page has fewer than N numeric words

Accepts one of the following:
int
str
page_contains_at_most_n_percent_numbers: Optional[Union[int, str, null]]

Trigger if page has fewer than N% numeric words

Accepts one of the following:
int
str
page_contains_at_most_n_tables: Optional[Union[int, str, null]]

Trigger if page has fewer than N tables

Accepts one of the following:
int
str
page_contains_at_most_n_words: Optional[Union[int, str, null]]

Trigger if page has fewer than N words

Accepts one of the following:
int
str
page_longer_than_n_chars: Optional[Union[int, str, null]]

Trigger if page has more than N characters

Accepts one of the following:
int
str
page_md_error: Optional[bool]

Trigger on pages with markdown extraction errors

page_shorter_than_n_chars: Optional[Union[int, str, null]]

Trigger if page has fewer than N characters

Accepts one of the following:
int
str
regexp_in_page: Optional[str]

Regex pattern to match in page content

regexp_in_page_mode: Optional[str]

Regex mode flags for regexp_in_page

table_in_page: Optional[bool]

Trigger if page contains a table

text_in_page: Optional[str]

Trigger if page text/markdown contains this string

trigger_mode: Optional[str]

How to combine multiple trigger conditions: 'and' (all conditions must match, this is the default) or 'or' (any single condition can trigger)

cost_optimizer: Optional[ProcessingOptionsCostOptimizer]

Cost optimizer configuration for reducing parsing costs on simpler pages.

When enabled, the parser analyzes each page and routes simpler pages to faster, cheaper processing while preserving quality for complex pages. Only works with 'agentic' or 'agentic_plus' tiers.

enable: Optional[bool]

Enable cost-optimized parsing. Routes simpler pages to faster processing while complex pages use full AI analysis. May reduce speed on some documents. IMPORTANT: Only available with 'agentic' or 'agentic_plus' tiers

disable_heuristics: Optional[bool]

Disable automatic heuristics including outlined table extraction and adaptive long table handling. Use when heuristics produce incorrect results

ignore: Optional[ProcessingOptionsIgnore]

Options for ignoring specific text types (diagonal, hidden, text in images)

ignore_diagonal_text: Optional[bool]

Skip text rotated at an angle (not horizontal/vertical). Useful for ignoring watermarks or decorative angled text

ignore_hidden_text: Optional[bool]

Skip text marked as hidden in the document structure. Some PDFs contain invisible text layers used for accessibility or search indexing

ignore_text_in_image: Optional[bool]

Skip OCR text extraction from embedded images. Use when images contain irrelevant text (watermarks, logos) that shouldn't be in the output

ocr_parameters: Optional[ProcessingOptionsOcrParameters]

OCR configuration including language detection settings

languages: Optional[List[ParsingLanguages]]

Languages to use for OCR text recognition. Specify multiple languages if document contains mixed-language content. Order matters - put primary language first. Example: ['en', 'es'] for English with Spanish

Accepts one of the following:
"af"
"az"
"bs"
"cs"
"cy"
"da"
"de"
"en"
"es"
"et"
"fr"
"ga"
"hr"
"hu"
"id"
"is"
"it"
"ku"
"la"
"lt"
"lv"
"mi"
"ms"
"mt"
"nl"
"no"
"oc"
"pi"
"pl"
"pt"
"ro"
"rs_latin"
"sk"
"sl"
"sq"
"sv"
"sw"
"tl"
"tr"
"uz"
"vi"
"ar"
"fa"
"ug"
"ur"
"bn"
"as"
"mni"
"ru"
"rs_cyrillic"
"be"
"bg"
"uk"
"mn"
"abq"
"ady"
"kbd"
"ava"
"dar"
"inh"
"che"
"lbe"
"lez"
"tab"
"tjk"
"hi"
"mr"
"ne"
"bh"
"mai"
"ang"
"bho"
"mah"
"sck"
"new"
"gom"
"sa"
"bgc"
"th"
"ch_sim"
"ch_tra"
"ja"
"ko"
"ta"
"te"
"kn"
specialized_chart_parsing: Optional[Literal["agentic_plus", "agentic", "efficient"]]

Enable AI-powered chart analysis. Modes: 'efficient' (fast, lower cost), 'agentic' (balanced), 'agentic_plus' (highest accuracy). Automatically enables extract_layout and precise_bounding_box when set

Accepts one of the following:
"agentic_plus"
"agentic"
"efficient"
source_url: Optional[str]

Public URL of the document to parse. Mutually exclusive with file_id

webhook_configurations: Optional[Iterable[WebhookConfiguration]]

Webhook endpoints for job status notifications. Multiple webhooks can be configured for different events or services

webhook_events: Optional[SequenceNotStr[str]]

Events that trigger this webhook. Options: 'parse.success' (job completed), 'parse.failure' (job failed), 'parse.partial' (some pages failed). If not specified, webhook fires for all events

webhook_headers: Optional[Dict[str, object]]

Custom HTTP headers to include in webhook requests. Use for authentication tokens or custom routing. Example: {'Authorization': 'Bearer xyz'}

webhook_url: Optional[str]

HTTPS URL to receive webhook POST requests. Must be publicly accessible

ReturnsExpand Collapse
class ParsingCreateResponse: …

A parse job.

id: str

Unique parse job identifier

project_id: str

Project this job belongs to

status: Literal["PENDING", "RUNNING", "COMPLETED", 2 more]

Current job status: PENDING, RUNNING, COMPLETED, FAILED, or CANCELLED

Accepts one of the following:
"PENDING"
"RUNNING"
"COMPLETED"
"FAILED"
"CANCELLED"
created_at: Optional[datetime]

Creation datetime

formatdate-time
error_message: Optional[str]

Error details when status is FAILED

name: Optional[str]

Optional display name for this parse job

tier: Optional[str]

Parsing tier used for this job

updated_at: Optional[datetime]

Update datetime

formatdate-time

Parse File

import os
from llama_cloud import LlamaCloud

client = LlamaCloud(
    api_key=os.environ.get("LLAMA_CLOUD_API_KEY"),  # This is the default and can be omitted
)
parsing = client.parsing.create(
    tier="fast",
    version="2025-12-11",
)
print(parsing.id)
{
  "id": "pjb-aaaaaaaa-bbbb-cccc-dddd-eeeeeeeeeeee",
  "project_id": "prj-aaaaaaaa-bbbb-cccc-dddd-eeeeeeeeeeee",
  "status": "PENDING",
  "created_at": "2019-12-27T18:11:19.117Z",
  "error_message": "error_message",
  "name": "Q4 Financial Report",
  "tier": "fast",
  "updated_at": "2019-12-27T18:11:19.117Z"
}
Returns Examples
{
  "id": "pjb-aaaaaaaa-bbbb-cccc-dddd-eeeeeeeeeeee",
  "project_id": "prj-aaaaaaaa-bbbb-cccc-dddd-eeeeeeeeeeee",
  "status": "PENDING",
  "created_at": "2019-12-27T18:11:19.117Z",
  "error_message": "error_message",
  "name": "Q4 Financial Report",
  "tier": "fast",
  "updated_at": "2019-12-27T18:11:19.117Z"
}