StayCharted
StayCharted AMT trains and tests text and image classification models using a team's labeled examples. Operations, support, finance, and software teams can use the resulting classifications in file workflows or, on eligible plans, through an API.
One of 75 tools in NoCode / LowCode
Who it's for
- Support and operations teams with labeled examples that they want to route into teams, categories, or priority bands.
- Finance, legal, and supplier-data teams that need to classify recurring business records using their own categories or decisions.
- Retail and inspection teams with categorized product or inspection photos who want to classify new images through files or an app.
Not the right fit if…
- Teams that need to read text inside receipts or forms from images: the Vision feature is not OCR.
- Teams that need automatic detection of names, street addresses, or free-text medical details in training data: the page says these are not detected yet.
- Free-workspace users who need API access: the page says API access is available on paid workspaces.
How it fits your workflow
Choose a recurring decision and provide labeled examples, such as tickets paired with categories or supplier records paired with approved matches. For images, provide a ZIP organized by category or a spreadsheet of picture links and labels.
Review the examples for label issues and, for text, inspect sensitive-data findings and choose whether to mask, exclude, or keep affected rows. Train a model and measure its results against examples it has not seen.
Use the model through batch files, file downloads, or an API where the plan provides access. Add examples and train an improved version as the business changes.
Pricing
Free: A lasting Free workspace includes one classifier model and file downloads. The page also states that sensitive-data checks are included on every plan for models that read text.
Free
Free
- · One classifier model
- · File downloads
- · Sensitive-data checks for text models
Essentials
Price not stated in supplied content
- · Classifier Models
- · AI Image Classifier Models available with the Vision add-on
Business
Price not stated in supplied content
- · AI Classifier Models
- · Dedicated AI Models
- · AI Image Classifier Models with Vision
- · API access is available on paid workspaces
Vision add-on
Price not stated in supplied content
- · AI Image Classifier Models on Essentials and Business
- · Dedicated AI Image Models on Business with Vision
Prices checked on Oct 6, 2026 from the vendor's site. They can change; confirm before you buy.
Key features
- No-code model training
- Train classification models using labeled examples without writing training code or needing machine-learning expertise.
- Text classification model options
- AMT offers Classifier Models for clearly worded text categories, AI Classifier Models that use meaning to handle varied wording, and Dedicated AI Models trained on a team's examples for more complex text distinctions.
- Image classification
- The Vision add-on supports training image classifiers from categorized photos or labeled picture links. New images can be categorized through files or an app using the model.
- Example review and testing
- Before training, users can review inconsistent labels and gaps; image workflows also identify duplicates, category mismatches, and unreadable links. Models can be tested against examples they have not seen.
- Sensitive-data checks for text
- Before training, AMT can flag specified sensitive values in text. Users can mask detected values, exclude affected rows, or keep them; models trained on masked text apply masking to new text inputs.
- File and API workflows
- Models can be used with batch files and file downloads, or through an API where the plan provides access.
Works with
- Prediction API
- Spreadsheets and batch files
- Existing business tools
Limitations to know
- Sensitive-data checks do not currently detect names, street addresses, or free-text details such as medical notes; those must be removed manually before training.
- The Vision feature classifies what a picture shows and is not OCR for reading text inside receipts or forms.
- Connecting the Prediction API to an application may require a developer.
- The page says API access is available on paid workspaces; the Free workspace is described with file downloads.
Alternatives to consider
-
categorAIze.io
Choose it if you want to categorize texts, URLs, images, or documents with LLM-powered categories without pretraining, rather than training models from your own labeled examples.
Getting started
- Pick a recurring decision with past examples and a clearly checkable correct outcome.
- Upload labeled text or spreadsheet examples; for images, use a category-folder ZIP or a spreadsheet of picture links and labels.
- Review the examples, train a model, and test it against unseen examples.
- Put the model into a file-based workflow or connect it to an application through the API where available.
Frequently asked questions
What happens to sensitive information in my training data?
AMT checks text before training so users can mask detected values, exclude affected rows, or keep them. Models trained on masked text also mask new inputs. The checks do not detect everything: names, street addresses, and free-text medical details need to be removed manually.
Can AMT learn from images?
Yes. The Vision add-on enables AI Image Classifier Models on Essentials and Business. Teams can train from categorized photos, then categorize new images through files or their own apps. Business with Vision also supports Dedicated AI Image Models. Vision identifies what a picture shows; it is not OCR for reading text inside receipts or forms.
What data do I need to get started?
For text, start with a spreadsheet or business-system export containing examples and their correct outcomes. For images, use a ZIP with one folder per category or a spreadsheet of public picture links and category labels.
Do I need coding or AI experience?
AMT provides no-code model training and does not require training code or machine-learning expertise. A developer may be needed to connect the Prediction API to an application.
Which model should I start with?
The page recommends a Classifier Model for clearly worded text categories, an AI Classifier Model when wording varies, an AI Image Classifier Model for pictures, and comparing a Dedicated AI Model using separate test examples when finer distinctions are needed.
Is my data private?
Models built by a user are private to that workspace. The shared, ready-made AI behind AI Classifier Models and AI Image Classifier Models does not learn from uploaded data. Dedicated AI Models are trained on the user's examples for their workspace.
Can I start for free?
Yes. The page describes a lasting Free workspace with Classifier Models and file downloads. Paid workspaces add more models, Dedicated AI Models, and API access.