How to cost-effectively use LLMs with Amazon Bedrock for PDF product manuals?
A company wants to use large language models (LLMs) with Amazon Bedrock to develop a chat interface for the company's product manuals. The manuals are stored as PDF files. Which solution meets these requirements MOST cost-effectively?
Community Votes
100% of anonymous learners picked answer D. Votes are pick records left by other test-takers — they are not the verified answer.
Community Insight
This question tests the ability to choose RAG (Knowledge Base) over prompt stuffing or fine-tuning when dealing with large, static document sets, balancing cost, accuracy, and scalability.
The most cost-effective way to build a chat interface over PDF product manuals with Amazon Bedrock is to use a Bedrock Knowledge Base, which leverages retrieval-augmented generation (RAG) to fetch only the relevant document chunks at query time.
Candidates often choose option A (adding one PDF via prompt engineering) because it sounds simple and avoids infrastructure, but embedding an entire PDF in every prompt wastes tokens and becomes far more expensive at scale than a knowledge base.
Community Discussion (4 comments)
Comments & Corrections
No comments yet — spotted an error or have a note? Share it below.
Expert Analysis
Why Option D is Correct
Amazon Bedrock Knowledge Base is a fully managed RAG (Retrieval-Augmented Generation) service. It ingests PDF documents, chunks them, generates embeddings, and stores them in a vector store. At query time, only the relevant chunks are retrieved and injected into the prompt. This means:
- You pay only for the input tokens actually used per query, not for entire manuals.
- No model retraining or fine-tuning costs.
- Easy to update when manuals change — just re-sync the knowledge base.
Why the Other Options Are Wrong
- Option A (one PDF as context): While cheaper than option B, stuffing an entire PDF into every prompt still wastes tokens on irrelevant content. Costs scale linearly with prompt size and are unpredictable.
- Option B (all PDFs as context): This is the most expensive approach. LLMs have context window limits, and sending all manuals per query will exceed them or incur massive token costs.
- Option C (fine-tuning): Fine-tuning in Bedrock is designed for teaching a model a new style or behavior, not for memorizing static documents. It incurs training costs, storage costs, and still doesn't guarantee factual recall of specific manual content. It is also overkill for this use case.
Key Takeaway
For question-answering over large document collections, RAG via Amazon Bedrock Knowledge Base is the AWS-recommended, most cost-effective pattern.
Official Reference
Exam Strategy
When an AWS AI question mentions large static document sets (PDFs, manuals, FAQs) and asks for cost-effectiveness, immediately think RAG / Knowledge Base — not fine-tuning, and never prompt stuffing.
Related Analysis
Practice All AIF-C01 Questions
Access 100 questions with complete answers and detailed explanations.
View Full AIF-C01 Practice Test →