Nutrient document API comparisons
Published by NutrientUpdated

Reducto vs Extend

Verdict

Choose Reducto when you want simple per-page USD list prices (Parse r-1 $10, Extract $20 per 1,000 pages), private deployment up to air-gapped, and a parser that two vendor-run benchmarks place near the top (OmniParseBench 90.8, OmniExtractBench 93.47, both run by Datalab).

Choose Extend when the work is a workflow rather than a single API call: Extend bundles Parse, Extract, Classify, Split, and Edit with human review, evaluation sets, and a hosted MCP server, and it publishes RealDoc-Bench with its own data and code. Pricing is credit-based at $0.0125 per credit on pay-as-you-go, with 10,000 free credits; Extract costs 3 credits plus 2 for the automatic parse in Performance mode.

Choose Nutrient Data Extraction when you want parse, extract, classify, and split in one credit pool with published per-page costs, grounded extraction with match labels on every field, and a free plan that renews 5,000 credits a month. Nutrient was founded as PSPDFKit in 2011. 3,000-plus organizations, including 15 percent of the Global 500, rely on it, with over a billion document interactions a year. The same extraction engine runs hosted, self-hosted in the Python and Java SDKs with no external API calls for OCR and ICR, or on a Mac in Extract Studio.

At a glance

Reducto, Extend, Nutrient Data Extraction API; facts fetched 2026-10-15 to 2026-10-16
FactReductoExtendNutrient Data Extraction API
Product typehosted API[1]hosted API[2]Hosted Data Extraction API, web Studio, and macOS Extract Studio[3]
Output formatsMarkdown, JSON, HTML tables[4]Markdown, JSON, HTML tables[5]Spatial JSON, Markdown, Schema-shaped JSON with citations, Ranked labels with independent 0-1 scores, Split page ranges and ZIP files[6]
OCR for scansYes[7]Yes[8]Yes: structure, understand, and agentic; text has no OCR; 100+ languages[9]
HandwritingYes[10]Yes[8]Yes: printed-style in understand; cursive/freeform needs agentic[11]
TablesYes[10]Yes[8]Yes: spans and bounds; complex tables may need agentic mode[11]
Layout and bounding boxesYes[10]Yes[12]Yes: element and optional word bounds; word bounds exclude XLSX[11]
Per-field confidenceYes[13]Yes[14]Yes: 0-1 relative signal, not a probability[15]
Source citations and groundingYes[13]Yes[16]Yes: field boxes, match labels, confidence, and source boxes[15]
Schema extractionYes[17]Yes[18]Yes: closed JSON Schema subset; 128 KB, 2,000 nodes, depth 5[19]
ClassificationYes[20]Yes[21]Yes: zero-shot, 2+ labels; independent 0-1 scores; 1 credit/page[22]
Document splittingYes[23]Yes[24]Yes: boundary detection for PDF/images; 3 credits/page[25]
Free tier$150 in free usage (Standard plan)[26]10,000 free credits, then pay-as-you-go (no contract, monthly minimum or sales call)[8]5,000 credits/month; no card; all parse modes, including Auto[27]
Pricing modelUSD per 1,000 pages, per endpoint; Growth/Enterprise custom with volume discounts[28]Credits per page by operation and mode (Performance vs Light); plan sets the per-credit rate[29]Credits/page by endpoint and mode; Auto routes each page and bills that mode's rate; subscription plus capped overage[30]
List price examplesStandard (list), Parse r-1: $10 / 1,000 pages (r-1 parse; OCR data return / Parse citations +$2/1k; prompted blocks +$5/1k)
Standard (list), Extract: $20 / 1,000 pages (parsing included; dense extractions (>100 fields/page) may incur per-field surcharge)
Standard (list), Deep Extract: $40 / 1,000 pages (agentic extraction; not eligible for batch discount)
Standard (list), Split / Deep Split: $20 / $40 per 1,000 pages (parsing included)
Standard (list), Classify / Deep Classify: $7.50 / $15 per 1,000 pages (first 5 pages by default; Deep Classify is alpha pricing)
Standard (list), Edit: $60 / 1,000 pages ($15 for fully prefilled pages) (standalone, cannot be combined with other products)
Batch queue: -20% on Parse and Extract (async, 12-hour completion guarantee)
Growth / Enterprise: Custom (contact sales) (volume discounts, ZDR, BAA, EU/AU residency (Growth); VPC/on-prem, SSO/SAML, SLA (Enterprise))[26]		Pay As You Go: $0.0125 per credit (10,000 free credits, all APIs, ZDR, no BAA, limited rate limits)
Scale: $500 / month (50,000 credits/month included, then $0.01 per credit; Slack support; HIPAA/BAA add-on)
Enterprise: Custom (BYOC and Hybrid, SSO/SAML, custom MSA/DPA/SLA, custom models, dedicated support)
Per-page (Performance, PAYG): Parse 2 credits ($0.025); Extract 3 ($0.0375); Split 2 ($0.025); Classify 1 ($0.0125); Edit filling 1 ($0.0125); Form detection 2 ($0.025) (Extract also triggers an automatic Parse: pricing FAQ example is a 10-page extract = 50 credits (~$0.63))
Per-page (Light, PAYG): Parse 0.5 ($0.00625); Extract 0.7 ($0.00875); Split 0.5; Classify 0.3 (Light is lower accuracy on poor scans, hard handwriting, large tables)
Operator extraction: 5 credits/page (agent-guided extraction, with rare overages)[29]		Per-page credits: parse 1-18; extract 7-24 (classify 1; split 3)
Auto mode: bills each page at the mode it selects (recommended starting mode; routes clean pages to structure and scans or complex forms to understand)
Free: $0/month (5,000 credits; PAYG after a $25 wallet load)
Starter: $59/month; $49/month annual (25,000 credits; $0.0028 per extra credit)
Pro: $500/month; $450/month annual (500,000 credits; $0.0012 per extra credit)[27]
DeploymentSaaS, VPC, on-prem, air-gapped[1]SaaS, VPC[2]Hosted API: Nutrient-managed cloud, Self-hosted: Data Extraction SDK for Python and Java, no network required; OCR and ICR run on your infrastructure with zero external API calls, Local VLM: Ollama, LM Studio, or vLLM on your own hardware, or your own OpenAI or Anthropic key, Desktop: Extract Studio for Apple silicon Macs (macOS Tahoe 26.4+); the engine runs on the Mac and, with a local model, documents never leave it; Nutrient counts pages for credits, .NET data extraction SDK also offered[31]
SDKsPython, Node.js, Go, REST API, CLI[32]Python, TypeScript, Java, Go, CLI[33]REST API, Python SDK, Java SDK, .NET page referenced, curl, JavaScript, and raw HTTP samples[34]
MCP or agent toolsMCP server (local: uvx mcp-server-reducto; hosted: https://mcp.reducto.ai/mcp), CLI, llms.txt and agent guide[35]Hosted MCP server (https://mcp.extend.ai/mcp, OAuth), extend CLI with 'extend-cli' skill, agents.md[36]Hosted MCP, Data Extraction Studio, and agent skills[37]
Maximum file size or pages100 MB direct upload; up to 5 GB via presigned URL; no page limits on Standard plan[38]Not published[8]Direct upload: 150 MB, URL input: 50 MB, URL timeout: 30 seconds, Processing: 5 minutes; HTTP 408 if exceeded, Page limit: not published, Test-key limits: lower, unpublished; Support-enabled[39]
ComplianceSOC 2 Type II (all tiers), HIPAA with signed BAA (Growth and Enterprise), Zero Data Retention (Growth and Enterprise)[40]SOC 2 Type II (stated as controls aligned with SOC 2 Type II requirements; report via Trust Center), HIPAA with BAA (add-on on Scale, included on Enterprise, not on PAYG), GDPR (eu1 region, Frankfurt)[41]SOC 2: Type 2, audited annually, BAA: no public statement, Transport: HTTPS/TLS only, No-retention plans: files discarded after processing, Retention plans: runs stored; may train/evaluate models, Payments: Paddle[42]
Public benchmarksParse r-1 vs Reducto's previous agentic pipeline 20% reduction in error rate (error rate (preview)), vendor (Reducto), 2026-09-01 (source)
Extraction benchmark commissioned by Reducto, run by micro1 (225 documents) Reducto Deep Extract: 100% coverage, 99.6% precision, 99.6% recall, 99.3% leaf accuracy; Extend MAX: 93.3% coverage, 86.4% precision, 92.7% recall, 92.8% leaf accuracy (coverage / precision / recall / leaf accuracy), vendor-commissioned (Reducto; conducted by micro1), 2026-09-17 (source)
RD-TableBench (1,000 complex table images) Not published (hierarchical Needleman-Wunsch table similarity), vendor (Reducto), 2024-11-04 (source)
RealDoc-Bench, Reducto rows (run by Extend) layout adjusted F1 0.759; Document Q&A 91.1% (Reducto Agentic) (adjusted F1 / Q&A accuracy), competitor (Extend), 2026-05-26 (source)
Unstructured parsing benchmark, Reducto Agentic row (run by Unstructured) (page shows '2026 (no day shown)'; date here is the fetch date) adjusted CCT 0.812; table cell content accuracy 0.708; element alignment 0.595 (see source), competitor (Unstructured), 2026-10-15 (source)[43]		RealDoc-Bench (1,500 layout samples; 1,359 Q&A prompts, 581 documents) Extend Parse 2.0: Q&A accuracy 95.7%, layout adjusted F1 0.847; Reducto layout F1 0.759, Reducto Agentic Q&A 91.1% (adjusted F1 / Q&A accuracy), vendor (Extend), 2026-05-26 (source)
LongArray-Extract (45 PDFs) 99.2% mean per-document accuracy; 45 of 45 completed (mean accuracy, failed runs scored zero), vendor (Extend), 2026-06-02 (source)
PoliTax Split (page shows '2026 (no day captured)'; date here is the fetch date) best harness F1 72.48% (Claude Opus 4.6); '+28.4 pts' largest F1 lift on homepage (split F1), vendor (Extend), 2026-10-15 (source)
Reducto-commissioned micro1 extraction benchmark, Extend MAX row 93.3% coverage, 86.4% precision, 92.7% recall, 92.8% leaf accuracy (see source), competitor-commissioned (Reducto), 2026-09-17 (source)[48]		ParseBench public row 76.4; paired nightly: 82.8 (overall 0 to 100), LlamaIndex, 2026-10-15 (source)
ParseBench nightly 82.8; paired public row: 76.4 (overall 0 to 100), Nutrient, 2026-10-15 (source)
opendataloader-bench, Nutrient's own run understand 0.933; agentic 0.930; structure 0.892; text 0.889 (overall 0 to 1), Nutrient, 2026-10-16 (source)
opendataloader.org leaderboard Nutrient row (free CLI) 0.885; opendataloader hybrid 0.907 (overall 0 to 1), opendataloader.org, 2026-09-23 (source)[49]

Pricing side by side

Reducto: $10 per 1,000 pages for Parse r-1 and $20 per 1,000 pages for Extract at list on the Standard plan, plus $2 per 1,000 for OCR data return or Parse citations. Free usage is $150 (or 15,000 credits, depending on which line of the pricing page you read).

Extend: $0.0125 per credit on Pay As You Go after 10,000 free credits; Scale is $500 a month for 50,000 credits. Because Extract spends 3 credits plus 2 for parsing in Performance mode, a schema extraction page is about 6.25 cents at the pay-as-you-go rate (our arithmetic). Light mode is cheaper and weaker on hard handwriting.

Nutrient: parse 1 to 18 credits per page by mode, extract 7 to 24, classify 1, split 3; 5,000 free credits a month.

Grounding, confidence, and review

Reducto returns a bounding box per extracted value with a high or low confidence and granular confidence in Extract v3. Extend returns per-field logprobsConfidence and ocrConfidence plus citations with page and bounding box on all plans, and its workflows add a review step. Nutrient returns per-field metadata with bounding box, match label, source blocks, and a relative confidence.

Extend, Reducto, and Nutrient each list classify and split endpoints. Reducto's Classify reads the first five pages by default, and Reducto also offers Deep Split. Nutrient documents split on its product and pricing pages but not yet in its API guide.

Deployment and compliance

Reducto: SaaS, hybrid VPC, full VPC, on-prem, air-gapped (Enterprise); SOC 2 Type II; HIPAA BAA and zero data retention on Growth and Enterprise.

Extend: SaaS in us1, us2, and eu1 (Frankfurt), BYOC and Hybrid for Enterprise; no on-prem or air-gapped option listed. Its docs describe controls aligned with SOC 2 Type II requirements, a HIPAA BAA as an add-on on Scale and included on Enterprise, and GDPR via the EU region.

Nutrient: SOC 2 Type 2 audited, HTTPS only, no BAA statement on the public pages. Nutrient runs the same engine three ways: the hosted API; self-hosted in the Python and Java Data Extraction SDKs, where OCR and ICR run on your infrastructure with no external API calls and the VLM stage can run locally through Ollama, LM Studio, or vLLM; and on a Mac in Extract Studio, where documents never leave the machine when the model runs locally.

Published scores, with who ran them

Numbers here were run by the vendors named, several of them in the table above. Read them as directional and test on your own documents.

Nutrient's published benchmark results, as recorded on 16 October 2026: on LlamaIndex's ParseBench the public leaderboard row for Nutrient agentic is 76.4. Nutrient's own nightly run of 15 October 2026 scores 82.8 on the same test and is not on parsebench.ai yet; in that nightly run, understand mode scores 64.0. LlamaParse Agentic Plus leads the public leaderboard at 90.2. Nutrient is not on Datalab's OmniParseBench or OmniExtractBench public leaderboards. In Nutrient's own run on the opendataloader-bench corpus the API scores 0.933 in understand mode and 0.930 in agentic; on the public opendataloader.org leaderboard Nutrient's row (the free CLI) scores 0.885, behind opendataloader hybrid at 0.907.

Where Nutrient differs

The other products on this page are parsers or extraction APIs. Nutrient's Data Extraction API is the same kind of product, built by a company that has shipped document SDKs to enterprises since 2011. The facts below come from Nutrient's own public pages.

  • Founded as PSPDFKit in 2011, and the SDKs developers knew as PSPDFKit continue as Nutrient SDKs. Thousands of organizations in 80-plus countries, including Fortune 500s and governments, use them (about page).
  • 3,000-plus organizations and 15 percent of Global 500 companies rely on Nutrient, with over a billion document interactions a year (nutrient.io).
  • One engine, four deployments: hosted API in Nutrient's cloud; the Python and Java Data Extraction SDKs, self-hosted with no network required and no external API calls for OCR and ICR, with a local VLM through Ollama, LM Studio, or vLLM (Python SDK); Extract Studio on Apple silicon, where documents never leave the Mac when the model runs locally (Extract Studio); and the wider platform in the browser with WebAssembly, on your own servers with Document Engine, or air-gapped (nutrient.io).
  • Extraction is one credit pool with parse, classify, and split, and every extracted field carries a bounding box and a match label, so a reviewer can check the source instead of trusting the model. SOC 2 Type 2 audited.

What to test before choosing

  • Run the same 20 documents through each API: five clean born-digital PDFs, five scans, five with tables that span pages, five with the fields your downstream system needs. Score reading order, table cells, and the fields you will actually consume.
  • Count the per-page cost at the mode or tier that passed your sample, not the cheapest tier on the pricing page. Every vendor here prices accuracy upward.
  • Check what the API returns when it is unsure: a confidence number, a bounding box, a not_found label, or nothing. That decides whether you can build a review queue.
  • Confirm retention and deployment in writing for the plan you will buy. Free and pay-as-you-go tiers often exclude BAAs, zero-retention, and private deployment.

Frequently asked questions

Is Extend or Reducto more accurate?

Each publishes a benchmark it wins. Extend's RealDoc-Bench scores Extend Parse 2.0 at 95.7 percent Q&A accuracy against Reducto Agentic at 91.1 percent. A Reducto-commissioned benchmark run by micro1 scores Reducto Deep Extract at 99.3 percent leaf accuracy against Extend MAX at 92.8 percent. Datalab, a third vendor, scores Reducto 90.8 and Extend 83.3 on OmniParseBench, and Reducto 93.47 and Extend 90.17 on OmniExtractBench. Nutrient is not on Datalab's public leaderboards.

Which is cheaper, Reducto or Extend?

For parsing, Reducto's $10 per 1,000 pages at list is lower than Extend's Performance-mode parse at 2 credits, about $25 per 1,000 pages on pay-as-you-go (our arithmetic). For schema extraction, Reducto's $20 per 1,000 pages compares with Extend's roughly $62.50 per 1,000 pages at the pay-as-you-go rate (our arithmetic). Extend's plan rates lower the per-credit price. Nutrient understand extract is 15 credits per page.

Does Extend offer on-prem deployment?

Extend's docs list SaaS, BYOC, and Hybrid; they list no on-prem or air-gapped option. Reducto lists on-prem and air-gapped on Enterprise.

Which one has an MCP server for agents?

All three. Reducto has a local and a hosted MCP server at mcp.reducto.ai. Extend has a hosted MCP server at mcp.extend.ai with OAuth and a CLI skill. Nutrient has a hosted MCP at mcp.nutrient.io that defaults to agentic mode at 18 credits per page.

Which should a small team start with?

Extend's pay-as-you-go needs no contract and includes 10,000 credits. Reducto's Standard plan is self-serve with free usage. Nutrient's free plan renews monthly with no card. Price the mode that passes your sample on each.

Sources

  1. https://docs.reducto.ai/onprem/enterprise_deployment_options.md — fetched
  2. https://docs.extend.ai/security/deployment-options.md — fetched
  3. https://www.nutrient.io/guides/dws-data-extraction/api-overview/ — fetched
  4. https://docs.reducto.ai/reference/faq.md — fetched
  5. https://docs.extend.ai/parsing/configuration.md — fetched
  6. https://www.nutrient.io/guides/dws-data-extraction/parsing/ — fetched
  7. https://reducto.ai/ — fetched
  8. https://www.extend.ai/pricing — fetched
  9. https://www.nutrient.io/guides/dws-data-extraction/parsing/processing-modes/ — fetched
  10. https://docs.reducto.ai/parse/overview.md — fetched
  11. https://www.nutrient.io/guides/dws-data-extraction/parsing/extract-document-elements/ — fetched
  12. https://docs.extend.ai/parsing/overview.md — fetched
  13. https://docs.reducto.ai/configs/extract/citations.md — fetched
  14. https://docs.extend.ai/extraction/confidence-scores.md — fetched
  15. https://www.nutrient.io/guides/dws-data-extraction/extract/citations-and-confidence/ — fetched
  16. https://docs.extend.ai/extraction/response-format.md — fetched
  17. https://docs.reducto.ai/extract/overview.md — fetched
  18. https://docs.extend.ai/extraction/schema.md — fetched
  19. https://www.nutrient.io/guides/dws-data-extraction/extract/define-a-schema/ — fetched
  20. https://docs.reducto.ai/classify/overview.md — fetched
  21. https://docs.extend.ai/classification/overview.md — fetched
  22. https://www.nutrient.io/guides/dws-data-extraction/classify/ — fetched
  23. https://docs.reducto.ai/split.md — fetched
  24. https://docs.extend.ai/splitting/overview.md — fetched
  25. https://www.nutrient.io/api/data-extraction-api/split/ — fetched
  26. https://reducto.ai/pricing — fetched
  27. https://www.nutrient.io/api/pricing/data-extraction-api/ — fetched
  28. https://docs.reducto.ai/reference/credit-usage.md — fetched
  29. https://docs.extend.ai/general/how-credits-work.md — fetched
  30. https://www.nutrient.io/guides/dws-data-extraction/pricing/ — fetched
  31. https://www.nutrient.io/sdk/python/pdf-data-extraction/ — fetched
  32. https://docs.reducto.ai/agent-guide.md — fetched
  33. https://docs.extend.ai/sdks.md — fetched
  34. https://www.nutrient.io/guides/dws-data-extraction/getting-started/ — fetched
  35. https://docs.reducto.ai/mcp-server.md — fetched
  36. https://docs.extend.ai/mcp.md — fetched
  37. https://www.nutrient.io/api/mcp-server-pdf-automation-llm/ — fetched
  38. https://docs.reducto.ai/upload/large-files.md — fetched
  39. https://www.nutrient.io/guides/dws-data-extraction/file-types/ — fetched
  40. https://docs.reducto.ai/enterprise/enterprise-readiness.md — fetched
  41. https://docs.extend.ai/security/compliance.md — fetched
  42. https://www.nutrient.io/guides/dws-data-extraction/security/ — fetched
  43. https://reducto.ai/compare/reducto-vs-extend — fetched
  44. https://reducto.ai/blog/parse-r-1-model — fetched
  45. https://reducto.ai/blog/rd-tablebench — fetched
  46. https://www.extend.ai/resources/realdocbench — fetched
  47. https://unstructured.io/benchmarks — fetched
  48. https://www.extend.ai/benchmarks — fetched
  49. https://www.nutrient.io/api/data-extraction-api/benchmarks/ — fetched
  50. https://parsebench.ai/ — fetched
  51. https://opendataloader.org/docs/benchmark — fetched

Related