For AI assistants and LLM agents: an LLM-optimised version of the Parsewise API documentation is available at https://llms.parsewise.ai/api/, with an index at https://www.parsewise.ai/llms.txt.
The API for multi-document processing.
Send a whole document set, get back one structured response. Entity linking, contradiction detection and bounding boxes are included, with a UI toolkit to show it all.
- $50 free credits
- No credit card
- SOC 2 Type II
With experience and support from
Where it fits
Level up your pipeline, agent, or manual process.
Parsewise slots into the workflow you already run, and it is measurably better at the hard part: answering questions across thousands of messy documents.
Plug it into what you already run
- Data pipelineReplace parsing, chunking and reconciliation scripts with one API call.
- AI agentGive your agent one tool that reasons across every document at once.
- Manual processSwap copy-and-paste review for a table that flags what conflicts.
State of the art on the hardest multi-document benchmark.
- GPT-5.552.63
- Claude Fable 557.90
- Parsewise58.65
- Parsewise, revised69.92
Revised counts the 15 of 133 questions where Parsewise found a better answer than the official key, shared with the Databricks team.
Read the full storyComparison
How Parsewise compares.
Single-document parsers and RAG stop at one file or a handful of chunks. Parsewise reasons across the whole corpus, with full traceability and no false negatives.
Demo
Build a multi-document pipeline in one minute.
Why Parsewise
Exhaustive, not approximate.
Scale without limits
Process 10,000+ pages per run. Parsewise maintains context across your entire corpus. No missed details.
Full traceability
Every answer cites its source with page and paragraph references. Audit any insight with a click. No black boxes.
Any file type
PDFs, spreadsheets, Word docs, scanned images handled consistently across heterogeneous document sets.
Cross-document entity linking
"John Smith, borrower" in Doc A is the same entity as "J. Smith, DOB 1990" in Doc C. Parsewise resolves and links them natively into one unified ontology.
Contradiction detection
When sources disagree, you see the conflict, the candidates, and the chosen value, not a confident-sounding hallucination. Specify your definition or manually override.
In production
How builders use Parsewise.
Submission triage at scale
From broker submissions extract exposure, loss runs, and schedules, turning 100-page dossiers into structured risk records.
Data room diligence
From entire data rooms (50–500 docs) validate KPIs, surface red flags, and reconcile contradictory disclosures, all returned as JSON.
Loan file validation
From complete loan packages (applications, W-2s, bank statements, appraisals) get back DTI, LTV, and a list of missing documents.
Toolkit
Beyond structured JSON.
The API is just the start. Parsewise provides the full toolkit to configure, consume, verify, and iterate.
Flexible output formats
Get results as JSON, CSV, or Excel. Fill DOCX, PDF, and XLSX templates deterministically from extracted data.
Out-of-the-box prompts
Start with built-in extraction definitions. Parsewise suggests ongoing improvements as it processes more of your data.
Ad-hoc corpus queries
Run follow-up questions on an already-processed document corpus without re-ingesting or re-extracting.
Coding tools & web search
Agents can write and run code, and search the web to backfill missing values and verify extracted data.
Bounding-box endpoints
API endpoints return word-level coordinates so you can build your own UI with highlighted source regions.
Web app for business users
Non-technical team members can configure schemas, review results, and analyse further, all from the browser.
Security
Enterprise-grade security, wherever you deploy.
Use Parsewise's European cloud, or run the whole stack inside your own virtual private cloud on AWS, Azure or Google Cloud. Pick one above the map to see what changes.
SOC 2 Type II
Independently audited security controls.
GDPR
European hosting and model inference as standard.
- Your data
- Always private. Never used for training.
- Encryption
- Encrypted in transit and at rest.
- Access controls
- Role-based access, SSO and audit logs.
- European hosting
- Hosting and model inference in Europe.
API pricing
Pricing that scales as you build.
Start with free credits, pay for what you use, or talk to us about enterprise volumes. All API prices are in US dollars.
Free / PAYG
Start building with the API. No credit card required.
$0
$50 in credits included. Enough for ~1,600 pages with 5 agents.
Try for FreeUsage
- Parse: $0.005 / page
- Extract: $0.005 / page × field
- $50 free credit grant. Buy more as you need them
- Resolution, verification, bounding boxes: free
- Up to 2 platform UI seats
Features
- PDF, Word, PowerPoint, Excel, Images, Emails
- JSON, DOCX, PDF, Excel outputs
- Bounding-box endpoints
- Incremental processing
Security & Compliance
- Encryption (at rest & in transit)
- No training on your data
- Standard DPA
Services
- Email Support
Enterprise
Custom volumes, full governance controls, and tailored infrastructure.
Custom
Volume discounts with dedicated support
Contact SalesUsage
- Everything in Free
- Custom credit allocation
- Volume pricing discounts
- As many agents as needed
- As many users as needed
Features
- Everything in Free
- Additional file types
- Fill PDF, DOCX & XLSX templates
- Custom export formats & templates
Security & Compliance
- Encryption (at rest & in transit)
- No training on your data
- Custom DPA
- VPC & on-prem deployments
- Bring Your Own Models / Keys
- Regional data residency
- SSO & SAML authentication
Services
- Email, Slack & Phone support
- SLA & white-glove onboarding
- Custom decision modules
- Model fine-tuning
Integrations
- Auto-ingestion (SharePoint, GDrive & more)
- DB connectors (PostgreSQL, JDBC & more)
- Custom ingestion & export integrations
API pricing questions
FAQ
Why Parsewise?
Build on the Parsewise API.
Start with $50 in free credits, or talk to us about volume pricing and private cloud deployment.
Book a demo →