THE SHORT ANSWER
Tonic Textual finds sensitive values in text, PDFs, Office files, scanned images and audio, then redacts them or swaps in consistent synthetic values, in Tonic's cloud or self-hosted. Alternatives include Presidio and Limina for de-identification, and GritWorks when you need offline redaction plus new synthetic documents and images generated with ground truth and degraded variants.
How this page was written. GritWorks makes GritRedact, GritTransform and GritRender, so we are not neutral. What we say about other products reflects their own public websites and documentation as of 17 November 2026. We make no claims about other products' accuracy. Products change: confirm details with each vendor.
What Tonic Textual does
According to Tonic.ai, Textual "finds sensitive information across your documents, transcripts, and files" and then redacts it or replaces it with synthetic values, using the same replacement every time a value appears so references stay consistent across a dataset.
- Inputs. The supported file types include .txt, .csv, .json, .docx, .xlsx, .pptx, .html, .pdf, .png, .jpg and .tif, and the product page adds audio. Tonic says Textual synthesizes PDFs while matching the original fonts, and detects logos and signatures in scanned forms.
- Deployment. Textual Cloud, or self-hosted in your environment with Kubernetes or Docker.
- Developer access. A Python SDK and a REST API, documented publicly.
- Pricing. A pay-as-you-go plan at a flat rate per 1,000 words processed, and custom Enterprise pricing.
Tonic also sells Fabricate, which generates relational data and free text that can be exported as PDF, DOCX or EML.
The useful question is not which tool handles PDFs (several do) but where your test data starts.
- You have real files and need them safe to use. De-identification tools such as Textual, Limina, Presidio and GritRedact start from your production documents, find sensitive values and remove or replace them. The output keeps the real variety of your data. It also inherits its gaps: rare document types, damaged scans and edge cases appear only as often as they did in production.
- You need documents that do not exist yet. Testing OCR, IDP and extraction usually needs specific layouts, controlled degradation (blur, fax noise, handwriting, stains) and a known correct answer for every field. That calls for generating documents from templates and rules, not transforming existing ones. GritRender generates documents, images, faces, fingerprints and audio with configurable fields and business rules, known values for scoring, and more than 30 negative document variants.
- Many teams need both. Redact a reviewed set of real documents, then use them as templates for generating larger, labelled variations. GritWorks supports that redact-then-generate path within one deployment, on-premises or as sales-assisted SaaS.
When to choose which
- Choose Tonic Textual when you want to make existing unstructured data usable for AI or development with realistic, consistent replacements, through an SDK and pay-as-you-go pricing, in Tonic's cloud or self-hosted.
- Choose Presidio when free, open-source text de-identification is enough and you will build the rest.
- Choose Limina when you want a de-identification API with a free plan and broad language coverage.
- Choose GritWorks when you are testing document AI (OCR, extraction, classification) and need layouts, scans, handwriting and degradation with a known answer per field, or when redaction must run offline with human review and an audit trail. GritTransform then extracts fields to JSON so you can score results against the known values.
To judge any synthetic document source, use the synthetic document data vendor checklist. The free sample packs show what labelled, degraded test documents look like.
What GritWorks does not replace
- GritWorks complements authorization, governance and e-discovery tools rather than replacing them. Access control, data catalogues, retention policy and legal review platforms keep their jobs; GritWorks prepares the documents, images and audio that move between them.
- Redaction is not authorization. If a person or system is not entitled to a record, block access instead of sending a redacted copy.
- Automated redaction alone is not a declaration of HIPAA de-identification. HIPAA recognizes the Safe Harbor and Expert Determination methods; policies, validation and human review must match the intended use and recipient.
- PII flagging is not redaction. GritTransform can flag potentially sensitive values in extracted results, but that does not remove them from the source. Use GritRedact when content must be removed.
- Engagement is sales-assisted. GritWorks has no public or self-serve API, SDK, MCP server or free tier, and publishes no price list: pricing is agreed with sales. Customer APIs are provided after onboarding.
Common questions
Is Tonic Textual text-only?
No. Tonic's documentation lists PDFs, PNG, JPEG and TIFF images, Word, Excel and PowerPoint files alongside plain text, and its product page adds audio. The difference from GritRender is that Textual transforms files you already have, while GritRender generates new documents from templates and rules.
What are the alternatives to Tonic Textual?
For de-identifying existing data: Presidio (open source), Limina (formerly Private AI), cloud PII services and GritRedact. For generating new labelled documents and images for OCR and extraction testing: GritRender. Tonic's own Fabricate generates relational data and free text exportable as PDF, DOCX or EML.
When is synthetic generation better than de-identifying real documents?
When the cases you need are rare or absent in production, such as damaged scans, handwriting, unusual layouts or specific business-rule failures, and when every field needs a known correct value to score against. De-identified real data keeps real variety but cannot add cases that never occurred.
Can GritWorks run inside our own environment?
Yes. GritRedact installs via Docker in your environment and runs fully offline with no network egress. GritRender also runs offline on your infrastructure, including air-gapped environments, or as sales-assisted SaaS.
Does GritWorks have a Python SDK or self-serve plan like Tonic Textual?
No. GritWorks has no public or self-serve API, SDK, MCP server or free tier. The web platform and APIs are provided to customers after sales-assisted onboarding, and pricing is agreed with sales.
How do I check that synthetic documents are realistic enough?
Run them through your own OCR or extraction pipeline alongside a small set of real, approved documents, and compare error patterns by field and condition. Synthetic data predicts real performance only when its structure and media characteristics are validated against the target workflow.