← All AWS exams

AWS · Professional · AIP-C01

AWS Certified Generative AI Developer - Professional

Independent practice, not affiliated with or endorsed by AWS. Study from the official exam guide.

The real exam

Questions
75
Duration
180 min
Passing score
750 (100–1000)
Practice bank
156 questions

Start practicing

Exam simulations, quick practice, and block-by-block study of the full bank.

Try 20 free questions · Log in to practice

Exam domains

  • Foundation Model Integration, Data Management, and Compliance (31%)
  • Implementation and Integration (26%)
  • AI Safety, Security, and Governance (20%)
  • Operational Efficiency and Optimization for GenAI Applications (12%)
  • Testing, Validation, and Troubleshooting (11%)

AIP-C01 practice exam and simulator

The AWS Certified Generative AI Developer – Professional (AIP-C01) exam tests whether you can put foundation models to work in real applications on AWS: retrieval-augmented generation, agents, prompt management, guardrails, cost and latency trade-offs, and troubleshooting when a GenAI feature misbehaves in production. Most questions are scenarios with several workable answers, and you have to pick the one that best fits the requirement in the last sentence.

Exam Ascent gives you a timed AIP-C01 exam simulator and a bank of practice questions organized around the exam's five domains, with an explanation for every answer option.

Try 20 questions free, no sign-up needed.

What the AIP-C01 exam covers

AWS's exam guide splits the scored content into five domains:

  • Foundation Model Integration, Data Management, and Compliance (31%): choosing and integrating foundation models, preparing data for retrieval, vector stores and knowledge bases, and meeting data-handling requirements.
  • Implementation and Integration (26%): building GenAI features into applications and workflows, including agentic solutions, prompt engineering and management, and connecting models to other services.
  • AI Safety, Security, and Governance (20%): guardrails, access control, protecting sensitive data, and Responsible AI practices.
  • Operational Efficiency and Optimization for GenAI Applications (12%): cost, latency and throughput: caching, model choice, and getting business value for the spend.
  • Testing, Validation, and Troubleshooting (11%): evaluating model output for quality and responsibility, monitoring, and fixing problems in production.

AWS expects candidates to have about two years of experience building production applications and one year of hands-on work with generative AI. Model training and advanced ML techniques are out of scope: this is a developer exam, not a data science one.

Exam format at a glance

  • Questions: 75 (65 scored, 10 unscored questions mixed in)
  • Time: 180 minutes
  • Question types: multiple choice (one correct answer) and multiple response (two or more correct answers; you must pick all of them)
  • Passing score: 750 on a scale of 100 to 1,000
  • Scoring: compensatory: you need to pass the exam overall, not each domain. Unanswered questions count as wrong, and there's no penalty for guessing.

Check the official AWS exam guide (linked above) for the full task list and the services in scope.

How practice works here

Exam simulation. A full-length set of questions drawn at random from the bank, timed at the real exam's pace. Mark questions for review and move around with the question navigator, as you would on test day. Your score and every explanation appear once you submit.

Study mode. No timer. Check each answer as you go and read the explanation straight away. Study sessions are kept out of your exam scores.

Quick practice and blocks. A short random set of 10 or 20 questions, or the whole bank in 50-question blocks.

Explanations for every option. Each explanation covers why the right answer fits the scenario and why each of the others doesn't. On AIP-C01, the wrong options are often real AWS features used in the wrong place, so knowing why they don't fit matters as much as knowing the right answer.

All-or-nothing scoring. A multiple-response question only counts when you pick every correct option and nothing else, the same as on the real exam.

Your progress. Every attempt is kept with your best, last and average scores, and progress charts show your score and time per question across tests, so you can see whether you're getting more accurate and faster.

Keep going if your connection drops. Answers are kept on your device and saved as soon as you're back online, even if you close the tab.

Tips for AIP-C01 questions

  1. Find the deciding requirement. "Least operational overhead", "lowest latency" and "most cost-effective" usually point to different answers for the same scenario.
  2. Prefer managed features over custom code when the question asks for the least effort, for example a managed knowledge base or guardrail over a hand-built pipeline.
  3. Watch for data boundaries. Questions about sensitive data often turn on where the data goes, who can see the prompts and responses, and what gets logged.
  4. Separate retrieval problems from model problems. Wrong answers from a RAG application may come from chunking, retrieval or prompts rather than the model itself.
  5. Take a full simulation early to check your pace: 75 questions in 180 minutes is about 2 minutes 20 seconds per question.

Exam details checked against the AWS Certified Generative AI Developer – Professional exam guide on 2026-10-19.

Frequently asked questions

How many questions are on the AIP-C01 exam?

75 questions: 65 are scored and 10 are unscored questions that AWS is trying out. The unscored questions aren't marked, so answer every question as if it counts.

How long is the AIP-C01 exam?

180 minutes, which is about 2 minutes 20 seconds per question.

What's the AIP-C01 passing score?

750 on a scale of 100 to 1,000. Scoring is compensatory, so you need to pass the exam overall rather than each domain.

Do I need to know how to train models for AIP-C01?

No. Model development and training, advanced ML techniques and feature engineering are out of scope. The exam is about building applications with foundation models: retrieval, agents, prompts, guardrails, cost and troubleshooting.

Can I practice for free?

Yes. Try 20 questions without creating an account. Your answers are scored in your browser, and you see an explanation for each question at the end.

How do I get full access?

Full access comes with an access pass for the exam. If your training provider has given you an invitation, use the link in your invitation email to sign in.

Other exams