FREE GENAI COURSES

Free Generative AI and Prompt Engineering Courses With Certificate, Compared

Google, Microsoft, Anthropic, Hugging Face, DeepLearning.AI and EDWartens, checked on their own pages: the course and the certificate are separate questions.

By EDWartens engineering team 26 September 2026 7 min
Free Generative AI and Prompt Engineering Courses With Certificate, Compared

A free generative AI course with certificate at no cost exists: Anthropic's AI Fluency course, the Hugging Face Agents Course and Google's Introduction to Generative AI badge all qualify. Microsoft's Generative AI for Beginners is free but offers no certificate. Google AI Essentials is paid. EDWartens teaches free, with an optional paid certificate.

Checked 26 September 2026. Every price and certificate term here comes from the provider's own page, linked under the post.

Free course or free certificate? Two different questions

"Free" on a course page can mean four things: the whole course is free, the videos are free but the labs are not, the first week is free, or the course is free and the certificate costs extra. Search results rarely tell you which. So we split every course into two answers: can you learn it without paying, and does proof of completion cost anything?

There is a third question worth asking: who issues the certificate, and can anyone check it? A badge on a provider's profile page, a PDF with a verification code and a proctored exam are three very different things. None of the free certificates here is a proctored certification exam, and none should be described as one on a CV. List what it is: the course name, the provider and "certificate of completion".

Free generative AI courses: course vs certificate
Free generative AI courses: course vs certificate

Google: one free course, one paid certificate

Introduction to Generative AI on Google Skills is the free one. Google Cloud Skills Boost now redirects to Google Skills. The Beginner: Introduction to Generative AI path has four activities, and its first course is a 45-minute introductory lesson on what generative AI is and how it differs from traditional machine learning. It earns a badge. Google Cloud's announcement of the Generative AI Fundamentals skill badge described it as no-cost.

Google AI Essentials is paid. Google lists it at US$49 a month in the US and Canada after a 7-day free trial on Coursera, with regional pricing elsewhere. It takes under five hours and ends in a certificate from Google. It is a good course for using AI at work, but it is not a free one.

Microsoft: free lessons, paid exams

Generative AI for Beginners is a 21-lesson course by Microsoft Cloud Advocates, published on GitHub under the MIT licence. Lessons are either "Learn" lessons on a concept or "Build" lessons with Python and TypeScript code. To run the code you need Azure OpenAI, the OpenAI API or Microsoft Foundry models, or Foundry Local, which runs models offline on your own device. The README lists no certificate.

Microsoft Learn training is free, and modules and learning paths earn badges and trophies on your Learn profile. The formal Microsoft certifications are exams with a fee set by country.

Anthropic: Claude Academy

Anthropic calls its learning hub Claude Academy and describes it as free courses, tutorials and use cases from its education team. The courses are hosted on Skilljar and marked free to register.

AI Fluency: Framework & Foundations was built with academic partners and covers delegation, description, discernment and diligence, including a deep dive on effective prompting. At the end you can take a final assessment and receive a certificate of completion. It needs no code, which makes it the easiest free prompt engineering course with certificate on this list.

Building with the Claude API is for developers: API calls, system prompts, structured output, prompt evaluation, tool use and retrieval-augmented generation. It is free to register, but the exercises start with getting an API key, so check the API's billing before you begin.

Hugging Face: two free courses

The Agents Course is free, and Hugging Face says its certification process is completely free. A fundamentals certificate follows Unit 1; a certificate of completion needs further units and assignments. It expects Python.

The LLM Course is free and without ads, covering Transformers, fine-tuning, datasets and tokenisers. It is the deepest free LLM course here, and the most technical.

DeepLearning.AI: free videos, paid labs and certificates

DeepLearning.AI's free plan gives access to all course videos. The paid Pro plan adds hands-on labs and the ability to earn professional certificates. Its catalogue is broad on prompting, RAG and agents, and the videos alone are worth watching. If a certificate is the point, budget for Pro.

EDWartens: generative AI for engineering documents

The EDWartens courses are aimed at engineers who will use language models on datasheets, manuals and reports:

  • [Generative AI and LLM Foundations](/free/generative-ai-and-llm-foundations) explains tokens and probabilities, how to choose a hosted or local model, how to call one from Python on a free route and run one on your own laptop with Ollama, and how to test a datasheet assistant before trusting it. Eight of its twelve modules need no code.
  • [Prompt Engineering for Engineers](/free/prompt-engineering-for-engineers) treats a prompt as a specification. Few-shot examples, structured JSON output, context engineering, prompts for FAT reports, I/O lists and procedures, and the habit of testing a prompt against a golden set before you rely on it.
  • [RAG and Chatbots with LangChain](/free/rag-and-chatbots-with-langchain) builds a chatbot that answers from PLC and drive manuals, cites the page and refuses what it cannot find.
  • [AI Agents and Workflow Automation with n8n](/free/ai-agents-and-automation-with-n8n) builds agents and workflows without code.

They run in Google Colab with free model routes. The AI foundations path puts the first two together with Python and n8n.

Free LLM course: which one explains how models work

"Free LLM course" can mean a course about using chatbots or a course about how the models work. For the second kind:

  • Hugging Face LLM Course. Transformers, tokenisers, datasets and fine-tuning, in code. Hugging Face designs each chapter for a week at 6 to 8 hours.
  • Microsoft Generative AI for Beginners. 21 lessons that move from concepts to building applications, in Python and TypeScript.
  • [Generative AI and LLM Foundations](/free/generative-ai-and-llm-foundations). Tokens in, probabilities out; choosing a hosted, local, open or free model; embeddings and vector search; function calling and agents; and when to prompt, use RAG or fine-tune.

Kaggle's Learn page also carries a self-paced 5-Day Intensive Generative AI guide made in partnership with Google.

What a good prompt engineering course makes you do

Most prompt engineering content is a list of prompts to copy. A course worth your time makes you do three harder things. First, write the task as a specification: the input, the output format, what to do when information is missing. Second, ask for structured output such as JSON or a table, so the answer can be checked by a script rather than by eye. Third, keep a golden set of inputs with known good outputs and re-run it whenever anything changes.

It should also cover the failure modes. A model given a confidential document can repeat it. A document can carry instructions the model follows. The EDWartens prompt engineering course ends its teaching modules on safety, confidentiality and injection for that reason.

How to choose

Check before you enrol
Check before you enrol

If you want a free certificate and no code, take Anthropic's AI Fluency course. If you write Python and want to build, take the Hugging Face Agents Course. If your job is engineering documents, take the EDWartens foundations and prompt engineering courses and decide on the certificate afterwards. Whatever you choose, the skill that matters is not writing a prompt. It is testing one: a fixed set of questions with known answers, run every time you change the prompt, the model or the documents.

Start free on EDWartens

Browse the free AI courses for engineers and create a free account so your progress is saved. Every course is free in full, with one final assessment of 15 questions, a 60% pass mark and three attempts. The optional EDWartens Certificate of Completion is a small one-off fee, from US$2.99 for a beginner course; checkout shows your price. Anyone can check it by code at edwartens.com/verification. It is not a vendor certification, a university award or an accredited qualification.

Take the free course

Questions

Which generative AI course has a free certificate?

Anthropic's AI Fluency: Framework & Foundations (free to register, certificate after a final assessment), the Hugging Face Agents Course (certification completely free) and Google's Introduction to Generative AI (a badge) all issue proof of completion at no charge.

Is Google AI Essentials free?

No. Google lists it at US$49 a month in the US and Canada after a 7-day free trial on Coursera, with prices varying by region. The certificate comes with the paid subscription.

Is there a free prompt engineering course with certificate?

Yes. Anthropic's AI Fluency course includes a deep dive on prompting and a free certificate of completion. Prompt Engineering for Engineers on edwartens.com is free to learn, with an optional paid certificate.

What is the best free LLM course?

For how models work and how to call them, Generative AI and LLM Foundations on edwartens.com or the Hugging Face LLM Course. For building agents, the Hugging Face Agents Course. For applications, Microsoft's Generative AI for Beginners.

Do I need a paid API key to learn generative AI?

Not always. The EDWartens courses use free model routes and Ollama on your own laptop. Microsoft's course lists Foundry Local for running models offline. Some provider courses call a paid API, so check the setup lesson first.

Sources

Written by the EDWartens engineering team for general education. Product names are trademarks of their owners; mentioning them does not imply endorsement. Prices and terms of other providers were checked on the date shown and can change.