My use case is that I want to be classify documents into one of several categories, then depending on the category, extract certain fields from it. My question is for this kind of use case, what exactly is the difference between using AI Skills to do this versus using AI Doc Center? Looking around, I see some people saying AI Doc Center uses more tokens and that it can handle more complex cases, but I can't see anything in documentation about if that's actually true or not, and anything about when you'd want to use one or the other. I've been in loops reading the same page over and over again, so at this point I might have missed something, but if someone's able to help point me towards an answer, that would be most appreciated!
Discussion posts and replies are publicly visible
Also, I want to add that I want to classify/extract about 500 documents a day
From what I've learned so far, if your requirement is simply to classify documents and then route the process based on the classification, AI Skills would likely be the better approach. It gives you the flexibility to classify the document and drive the workflow accordingly.
If your use case goes beyond classification—for example, extracting different fields from multiple document types, handling confidence scores, review workflows, or managing document extraction models—then AI Document Center is probably the better fit since it's designed specifically for intelligent document processing.
Regarding token usage, I've seen community members mention that AI Skills may consume fewer tokens for simpler scenarios, but I haven't found any official Appian documentation that confirms this.
AI Skills are the runtime objects. You build them directly pick classification or extraction, train or prompt them, and call them from a process via smart services like Classify Document and Extract from Document.Doc Center is not a separate engine. It's an application that generates and maintains those same AI skill objects and their process models for you, then layers on versioning, reconciliation, accuracy metrics, token monitoring, and deployment.Both end up as AI skill objects at runtime. The choice is whether you build and manage the AI skill yourself, or let Doc Center generate it and give you versioning and accuracy metrics. Doc Center's models are gen-AI backed (tokens, better on varied layouts); an AI skill you build directly can be trained ML: no tokens, but needs consistent layouts and retraining.
Regarding extraction, I just need to be able to handle the following situation - if it's doc type A, then extract XYZ fields. if it's doc type B, then extract ABC fields. Sounds like then I'd probably want AI Skills to classify which document type it is, then use AI Doc Center to extract the values?
Also, when you say 'review workflows or managing document extraction models' for AI Doc Center, what particular features are you referring to? What about workflows can you review in the AI Doc Center, and what can you manage regarding document extraction that you can't do in AI Skills?
Where exactly can you monitor tokens for Doc Center? In Doc Center, I can see Home, Classification, Extraction, Deployment, Documentation tabs on the left side, but nothing about token monitoring?
Also, where in the documentation does it say that AI Skills use no tokens? Want to just have a link as reference
Use MyAppian -> Adoption -> AI Usage Metrics usage to see AI Action consumption by feature across environments.Also from 26.6 in logs also :- https://docs.appian.com/suite/help/26.6/Logging.html#ai-usage-logs
If your documents are already well-defined document types and your requirement is simply "if it's document type A, extract fields XYZ; if it's document type B, extract fields ABC," then I don't think I'd split the solution between AI Skills and AI Document Center. AI Document Center is capable of handling both the classification and the subsequent field extraction based on the identified document type.
The scenario where I'd lean towards AI Skills is when you only need classification (or some custom AI reasoning) and don't need the document-processing capabilities that AI Document Center provides.
kl0001 said:Also, when you say 'review workflows or managing document extraction models' for AI Doc Center, what particular features are you referring to? What about workflows can you review in the AI Doc Center, and what can you manage regarding document extraction that you can't do in AI Skills?
Lets say you are trying to extract some fields and confidentiality score is low in that case you can review that and proceed further that is what I mean by review workflow https://docs.appian.com/suite/help/26.6/aidc-4.2/reconcile-extraction.html
Hi Shubham, I can see input/output token usage for unstructured doc extraction ai skills and for doc center, but i don't see it for something like structured doc extraction ai skill. Do you know how I can view that?