Sold by: Software Sushi
Deployed on AWS
Free Trial
Offline identifier and email redaction for Spanish documents from five Latin American countries. Masking, typed replacement, keyed pseudonyms and detection-only modes. Names and addresses are excluded.
Overview
LatAm Spanish PII Redaction 0.2.0 - identifiers-only release
Version 0.2.0. The SageMaker model package is certified for Marketplace. Publication and commercial terms were approved by the product owner; consult the Marketplace offer for current availability.
Detect and redact identifier candidates in Spanish text from Mexico, Chile, Argentina, the Dominican Republic and Colombia, plus email-address syntax. The eight country-specific types are AR_CUIT (CUIT/CUIL), AR_DNI, CL_RUT (RUT/RUN), CO_CEDULA, CO_NIT, DO_CEDULA, MX_CURP and MX_RFC; email matches use EMAIL.
Choose character masking, typed replacement, server-keyed pseudonyms or detection-only metadata. JSON, JSON Lines, plain text and strict one-column CSV support realtime and batch workflows. Spans use Unicode code-point offsets in unchanged input. Entity metadata does not contain raw matched values.
PERSON and ADDRESS detection are unavailable. Names, addresses and other unmatched content can remain in the output. The release profile is locked to identifiers/email; this is not complete anonymization.
Detection and validation differ. Recognized failed, masked, generic and unverified candidates remain redacted. AMBIGUOUS_ID is a protective, unverified emission when nationality is unresolved, not another validated national scheme. CURP and Dominican cedula checksums are unverified; RFC arithmetic is partial and generic or ambiguous cases remain unknown. No result verifies issuer registration, identity, mailbox ownership or legal compliance.
An independently authored, post-freeze AI-generated V4 set produced 600/600 exact document results, with zero false positives, false negatives or expected-value leaks across mask/replace/hash. These frame-based synthetic documents had no human review and do not establish population accuracy. See the model card for measured image/AWS results, earlier failed evaluations and evaluation limitations.
The approved software price is $0.50 per instance-hour on either ml.m5.large or ml.c5.xlarge, with a 14-day software trial. AWS infrastructure charges are separate. Final Marketplace terms govern. Subscribe through AWS Marketplace and deploy with SageMaker, then follow usage or the synthetic notebook with your own endpoint. No private repository or private image access is required. Review identifier sources and license boundaries .
Watch the product workflow
See supported identifiers and email addresses processed by the actual published model, with typed replacement, character masking and explicit validation limits. Names and addresses remain outside this release.
Watch the English product video on YouTube . The interface is an illustrative integration rather than an included product UI. The examples are fictional; model responses are real.
Highlights
- Offline deterministic identifier and email detection with explicit validation limits
- Mask, typed replacement, keyed pseudonyms and detection-only modes
- Runs in your AWS account with no runtime downloads; names and addresses excluded
Details
Sold by
Categories
Delivery method
Latest version
Deployed on AWS
New
Introducing multi-product solutions
You can now purchase comprehensive solutions tailored to use cases and industries.
Features and programs
Financing for AWS Marketplace purchases
AWS Marketplace now accepts line of credit payments through the PNC Vendor Finance program. This program is available to select AWS customers in the US, excluding NV, NC, ND, TN, & VT.
Pricing
Free trial
Try this product free for 14 days according to the free trial terms set by the vendor.
Pricing is based on actual usage, with charges varying according to how much you consume. Subscriptions have no end date and may be canceled any time.
Additional AWS infrastructure costs may apply. Use the AWS Pricing Calculator to estimate your infrastructure costs.
Dimension | Description | Cost/host/hour |
|---|---|---|
ml.m5.large Inference (Batch) Recommended | Model inference on the ml.m5.large instance type, batch mode | $0.50 |
ml.c5.xlarge Inference (Real-Time) Recommended | Model inference on the ml.c5.xlarge instance type, real-time mode | $0.50 |
ml.m5.large Inference (Real-Time) | Model inference on the ml.m5.large instance type, real-time mode | $0.50 |
Dimensions summary
You pay by the hour for each running instance, with no upfront commitment. Pricing follows two variables: the EC2 instance type and the inference mode. Two options run on the ml.m5.large instance, split between real-time mode and batch mode. A third option runs on the ml.c5.xlarge instance in real-time mode. Charges scale with how many hours each instance runs. AWS infrastructure, data transfer, and any connected third-party services are billed separately from this software charge.
Top-of-mind questions for buyers
Each option bills per running host-hour on a specific EC2 instance type. The ml.m5.large and ml.c5.xlarge differ in compute and memory resources. The instance type you pick sets the hourly software rate, so faster or larger compute carries a different per-hour charge.
Real-time mode runs a live endpoint that responds to requests as they arrive, so you pay for the hours the endpoint stays up. Batch mode processes a set of inputs in one job, so you pay for the hours that job runs. Both bill per host-hour.
The hourly charge covers the redaction software only. AWS infrastructure such as EC2 compute, storage, and data transfer is billed separately. If you connect remote detectors or other external services, those providers may charge their own usage fees under their own accounts.
softwaresushi.com
Helpful?
Vendor refund policy
Software charges stop when the endpoint is deleted or a batch job ends. If the product did not work as described, email info@softwaresushi.com within 30 days with your AWS account ID and endpoint or job name. Software Sushi will refund the software charges for that use. AWS infrastructure charges are separate and are not covered.
Legal
Vendor terms and conditions
Upon subscribing to this product, you must acknowledge and agree to the terms and conditions outlined in the vendor's End User License Agreement (EULA) .
Content disclaimer
Vendors are responsible for their product descriptions and other product content. AWS does not warrant that vendors' product descriptions or other product content are accurate, complete, reliable, current, or error-free.
Delivery details
Amazon SageMaker model
An Amazon SageMaker model package is a pre-trained machine learning model ready to use without additional training. Use the model package to create a model on Amazon SageMaker for real-time inference or batch processing. Amazon SageMaker is a fully managed platform for building, training, and deploying machine learning models at scale.
Deploy the model on Amazon SageMaker AI using the following options:
Version release notes
Version 0.2.0 identifiers-only release. Eight national identifier types plus email and protective ambiguous-ID detection. Names and addresses are excluded. Immutable offline profile; four redaction/detection modes. Partial and unknown checksum coverage is documented. Independent AI-generated synthetic evaluation and finite AWS measurements are disclosed in the model card; no representative customer-accuracy or latency-SLA claim.
Additional details
Inputs
- Summary
UTF-8 Spanish text as JSON, JSON Lines, plain text or one-column CSV with a text header.
- Limitations for input type
- 256 KiB body; 16 records; 16000 Unicode code points per record; 64000 total per request. Names and addresses are not detected in this release.
- Input MIME type
- application/json, application/jsonlines, text/plain, text/csv
{"mode": "replace", "text": "RFC LUNA910210AB1; RUT 12.345.678-5"}
{"mode": "replace", "text": "DNI 23456789"}
{"mode": "mask", "text": "CC 1234567890"}
Input data descriptions
The following table describes supported input data fields for real-time inference and batch transform.
Field name | Description | Constraints | Required |
|---|---|---|---|
text | Text to inspect. | At most 16000 Unicode code points in the rules profile. | Yes |
mode | mask, replace, hash or detect_only. | Hash requires a server-managed secret key; mask is the default. | No |
country_hint | auto, MX, CL, AR, DO or CO. | Hints allow ambiguous bare identifiers and increase false positives. | No |
Resources
Support
Vendor support
For product support, contact info@softwaresushi.com . Review the model card and documented detection and validation limits before processing permitted personal data.
AWS infrastructure support
AWS Support is a one-on-one, fast-response support channel that is staffed 24x7x365 with experienced and technical support engineers. The service helps customers of all sizes and technical abilities to successfully utilize the products and features provided by Amazon Web Services.
Similar products
This solution enables users to custom define the sensitive information and automatically redact them using LLM based in-context reasoning.
Veritone Redact, powered by aiWARE, performs automated head, object, and license plate detection and
redaction on audio and video files. Remove costly manual processes and reduce
human error with our AI-powered SaaS solution for in-house redaction teams, or our redaction on-demand offering performed by our team of redaction professionals.
Mask or Obfuscate Personal Health Information (PHI) in English clinical notes.
Slalom's intelligent document processing redaction accelerator provides an out of the box user interface to upload large volumes of documents, with PHI & PII redaction and archiving capabilities that are modular and scalable to many AI use cases, saving both time and money while enabling faster responses to constituent and customer requests.
Customer reviews
No customer reviews yet
Be the first to review this product . We've partnered with PeerSpot to gather customer feedback. You can share your experience by writing or recording a review, or scheduling a call with a PeerSpot analyst.