AI with Fable-level Performance, Working for the Price of a Coffee - Kimi K3 Vibe Coding
"The cost of making one app is the price of a cup of coffee."
Learn vibe coding with Kimi K3, an open-weight model with 2.8 trillion parameters.
You’ll launch your first app in a browser with a single sentence, create a UI by providing a hand-drawn sketch, and have it build and deploy an entire full-stack web app to the internet.
If you’re already using Claude Code, you’ll also learn how to keep that environment as-is and simply switch the model to K3.
It’s pay-as-you-go, not a monthly subscription. In an actual test, one session that created and deployed an app cost 1.69 dollars.
14 lessons, approximately 102 minutes total; the course is free.
The Vibe-Coding Skill of Launching Browser Apps with a Single Prompt (Master It Through 6 Hands-On Exercises)
A work environment with Kimi Code CLI installed, an API key issued, and a small balance topped up (Windows and macOS)
The vibe-coding loop of requesting, checking, and having it fixed — how to write prompts using four sections: goals, scope, criteria, and constraints
A vision workflow that lets you create and modify a UI by providing a hand-drawn sketch or a single screenshot.
Deploy the app you built to Vercel and share it via a link — a real URL with my name on it
Project-scoped setup for integrating K3 into your existing Claude Code without affecting your usual environment
How to Use a 1M Context to Complete a Long, Two- to Three-Hour Task from Start to Finish Without Summarization (Compaction)
If you want to build something with AI, do you have to pay for a subscription first?
When you look into AI coding tools, the first hurdle is a subscription. $20 a month, $100 if you want to use it more, or $200.
But after you pay, months like this come along. I used it three times this month.
There’s the opposite, too. You want to spend one day over the weekend building something, but you pay for a full month just for that one day.
There is an option to pay only for what you use. It’s pay-as-you-go.
And in July 2026, one more option was added to the mix. The 2.8-trillion-parameter open model, Kimi K3.
The weights are publicly available, it has a context window of 1 million tokens, and it reads images and videos as they are.
The official blog lists frontier-level models as this model’s comparators.
And the pricing is $3 per 1 million input tokens, $15 per 1 million output tokens, and $0.30 per 1 million cached input tokens.
This course is about actually having that model do real work.
One app costs a cup of coffee
14 lessons, approximately 102 minutes. The course is free.
The first three sessions are for setup. It takes 19 minutes to obtain an API key, add a small amount of credit, install the Kimi Code CLI, and launch your first app in a browser immediately after installation.
From there, we get to the main point.
You’ll learn the loop of making requests, checking the results, and having them fixed; create a UI from a single hand-drawn sketch; and have a full-stack web app built and put online so you can share it via a link.
Complete the entire course with a single API key. There is no subscription.
Use the same key with Kimi Code and when connecting directly to Claude Code.
What you type is a request. The AI writes the code.
What You’ll Have at the End of the Course
① First app - A coffee cost converter made with a single sentence. It runs in your browser within 10 minutes of installation.
② UI from a hand-drawn sketch - a screen created by submitting a single design sketch drawn on paper
③ Full-stack web app - A token cost calculator consisting of a frontend and server API. From planning to validation
④ My URL - The actual address you get after deploying ③ to Vercel. It can be shared as a link.
⑤ Claude Code with K3 embedded - A project folder with the same environment you were using, only with the model changed
⑥ Cost awareness - The final piece: enter the tokens you used into ③ to calculate the actual cost.
The six parts do not stand alone. ① is a scaled-down version of ③, ③ is deployed as ④, and finally, ③ is used to do ⑥.
Three reasons to choose this course
1. Not how to use a tool, but how to “get work done while paying as you use it”
This is not a course about memorizing commands.
It covers how to write requests (goals, scope, criteria, and constraints), how to verify the results, and how to decide what to delegate and where to draw the line.
This sense of judgment remains even when the model changes.
2. It Covers Both Paths
We cover both paths: going from start to finish with Kimi Code CLI (Parts 1 and 2), and connecting K3 directly to Claude Code you are already using (Part 3).
In Lesson 11, we have both tools perform the same tasks side by side and compare their speed, output, and cost in a split-screen view.
K3 is the model on both sides. You will be able to distinguish what differences come from the tools and what differences come from the model.
3. We Only Discuss Measured Results
The costs mentioned in this course are amounts the instructor actually paid, and we show the console screen exactly as it is.
We also point out when it costs more than a cup of coffee. If you run it all day, every day, it costs more than a cup of coffee. It’s usage-based pricing.
We also point out which features don’t work. When connected to Claude Code, the web-reading feature doesn’t work, and account integration tools are disabled.
Lesson 14 is devoted entirely to discussing the limitations.
① Your First App with a Single Phrase — 10 Minutes Right After Installation
Once the installation is complete, start building right away. There’s no syntax explanation or code to type along with.
Request
"Make me a single-file webpage that shows how many cups of coffee the amount I spent equals."
Kimi Code creates the file, saves it, and opens it in the browser.
Enter a number, and the number of cups appears. This app will be brought back for redesign in Lesson 5,
then rebuilt as the official version in Lessons 6 and 7.
The goal of this session is for you to see a “working result made by giving verbal instructions” within the first 10 minutes.
② Toss in a hand-drawn sketch and have it make the UI
K3 reads images as they are. No separate tools need to be added.
Request
"Rebuild the screen according to this sketch. Look at sketch.jpg."
A hand-drawn sketch on paper, a screenshot of a site you like, or an arrow roughly marked with a mouse.
Instead of saying something like “Make this button a little bigger,” you simply provide the screen itself.
The key to this lesson is the ability to convey requests that are difficult to explain in words through images.
③ Have It Build the Entire Full-Stack App (2 consecutive sessions)
Combine sessions 6 and 7 to create an app from planning through validation.
What to Build – Token Cost Calculator
Enter the input, output, and cache tokens to see a cost table by model, with the lowest price highlighted and the total converted into the number of cups of coffee.
Calculation history is saved in the browser and remains in a list.
Stack Next.js + server API Route. No database is used.
(Why we don't use one and when it becomes necessary are covered in Session 14.)
Session 6 covers everything from making a request → reviewing the plan → scaffolding → implementing features,
and Session 7 is the loop of completing the app → testing it by actually clicking through it → having it fixed.
We show you the actual defects that emerge along the way and include the process of having them fixed.
④ Deployment – A Result That Can Be Shared via a Link
We upload the app we made to Vercel. We also have the agent handle the deployment.
There is a noticeable difference between something that only runs locally and something that has its own URL.
It opens on your phone, you can send the link, and when you make changes, it gets redeployed.
We’ll use this address again in Lesson 13. It’s for calculating your own costs with the calculator you created.
⑤ Embedding K3 into the Claude Code You’ve Been Using (Part 3)
If you are already using Claude Code, this part will probably be the most practical for you.
With the Anthropic-compatible endpoint and Kimi API key set as environment variables, K3 runs using the workflow you’re already familiar with.
In Lesson 9, you’ll configure 10 lines and verify them; in Lesson 10, you’ll do hands-on work; and in Lesson 11, you’ll have a head-to-head showdown with Kimi Code.
Keep the configuration only inside the project folder. If you put it in the global configuration, your usual Claude Code setup will be changed entirely.
We’ll show you how to keep using it as usual outside that folder. We’ll also cover how to revert the changes.
Notice: This setup is not officially supported by Anthropic.
Anthropic’s documentation states that routing Claude Code to non-Anthropic models is not supported, and its behavior may change when the client is updated.
This course uses only a Kimi API key without using Anthropic subscription credentials.
Additionally, web reading does not work with this setup, and account-linked tools (connectors) are disabled.
We will confirm these limitations in Lesson 10 and also cover ways to work around them.
⑥ Complete Long Tasks from Start to Finish Without Summarization – 1M Context Window
When you work with AI for a long time, the context window fills up. That’s because every file it has read, every exchange, and every result produced by tools keeps accumulating.
Once it’s full, the tool summarizes and discards the earlier parts (compaction). At that moment, it pauses for several minutes, then continues with a summary missing the details, rereading the same files or forgetting decisions made earlier and starting to build things differently. This problem doesn’t show up in short app tasks, but emerges in work sessions lasting two or three hours.
K3 has a context window of 1 million tokens. The price per unit does not increase even as the context gets longer. It is the same throughout.
In Session 12, we put both implementations from the head-to-head comparison in Session 11 (340 files, 310,000 tokens of code alone) into a single context window and had them conduct a specification-by-specification comparative review.
We confirm on screen that, even as the gauge passes the 200K mark, summarization never occurs.
In actual measurements, it reached a maximum of 338,000 tokens, with 0 compactions, taking 23 minutes and costing $5.11.
What 1M provides is not so much “putting more in” as “not losing your memory during long tasks.”
Is it really the price of a cup of coffee? Let’s look at the numbers laid bare.
Since I wrote “coffee money” in the title, I’ll lay out the basis here.
The bill is made up of three lines (per 1 million tokens)
Item
Unit price
What accumulates?
Input
$3.00
The requests I sent and the files the AI read
Output
$15.00
Code, explanations, and thoughts written by the AI
Cached input
$0.30
When rereading something that has already been read (1/10 of the input)
Coding work involves repeatedly looking at the same files, so in the actual bill, the third line makes up the largest share.
That’s why continuing to work within a single session is cheaper than turning it off and on repeatedly.
Actual costs - What I really spent creating the course
Sessions 6–8, one session to build and deploy an app from scratch: $1.69 (about 2,400 won, half a cup of coffee)
Average across 7 recording sessions: $1.41. The most expensive one was $5.11 for having it read through two entire codebases in Session 12.
24 web app sessions, including the additional experiments run while creating the course in the recording sessions: a total of $122.49, an average of $5.10 per session (about 7,100 won, the price of a premium cup of coffee)
These aren’t made-up numbers; they’re the values shown in the console and session logs, and in Lesson 13, I show you the console screen exactly as it is.
And let me tell you about the other side as well
Building one app does cost about as much as a cup of coffee. I say this after running it multiple times.
However, that’s not the case if you run it all day, every day. Since it’s pay-as-you-go, you pay for what you use.
More precisely, "one app costs one cup of coffee," not "one cup of coffee per month."
Then how much would it cost in my case? Enter your own usage into the calculator you’ll build in Lessons 6 and 7. We’ll cover how to do that in Lesson 13.
※ The unit prices are based on the rates at the time this page was written. Pricing policies may change, so please check platform.kimi.ai again before making a payment.
This is a sponsored course.
This course was created with support from Moonshot AI. I thought it would be better to disclose that upfront.
So there are three principles I adhered to even more strictly.
We use only measured costs. We show the console screen exactly as it is and also mention the periods when the cost is more than just the price of coffee.
We did not leave out the things that don't work. Lesson 14 is entirely devoted to limitations.
We have organized five issues, along with their symptoms and solutions: taking too long even for minor edits because it is always thinking, touching files it was not asked to modify, being pulled along by the previous direction when the conversation gets long, being unable to read the web when integrated with Claude Code, and account-linked tools being disabled.
I'm not telling you to abandon the tools you've been using. The instructor hasn't abandoned them either.
In Lesson 13, I'll explain when to choose K3 and when to choose other models.
If you go through them without understanding, you may give up, but if you know what to expect, you can get past them. That’s why I included them all.
Preparation - All you need is one key
Sign up for platform.kimi.ai - We’ll go through it together while viewing the screen in Lesson 2.
Issue an API key - Create a key in the console
Small amount of credit - You can add credit starting from just $1 (charged in USD). Once your cumulative top-up reaches $10, the per-minute request limit increases from 3 to 100, so we recommend $10 to reduce waiting during practice.
Install the Kimi Code CLI - macOS and Linux require one command, while Windows requires one PowerShell command.
Authentication - Paste your API key at /login and check it with /status—that’s it.
You will use this one key throughout the entire course. It is the same key for Kimi Code and when connecting directly to Claude Code in Part 3.
The only payment required for the course is this credit top-up.
Student Benefit - 15% Additional Credits
If you sign up as a new user through the link below, you will receive benefits when purchasing credits.
This applies only to accounts that sign up through the link above.
When purchasing API credits, an additional 15% in credits is provided with each order.
Limited to orders placed by December 31, 2026
Bonus credits will be added to your account the day after your purchase.
If you already have a Kimi account, you are not eligible. This benefit applies only to newly registered accounts.
It will not apply if you click this link again using an existing account. We are stating this explicitly so that you are aware in advance.
※ There is no commission paid to the instructor when you use this link. This is a benefit provided by Moonshot AI to students taking this course.
What about the membership?
The current membership is operated on an application-based waiting list.
If you’re interested, you can apply, and please check Kimi’s official announcement for details.
This course uses an API pay-as-you-go model from start to finish. If you’re starting now, this is the option for you.
Recommended for those who:
✅ Those who want to try vibe coding but have been putting it off because the monthly subscription fee feels burdensome
✅ Those who subscribe to AI coding tools but feel it’s a waste because they only use them two or three times a week
✅ Planners, marketers, and operators with no programming experience who want to build the tools they need themselves
✅ Developers who want to see for themselves whether an open model with publicly available weights is practical for real-world use
✅ Those who are already using Claude Code and want to try changing only the model while keeping their existing environment unchanged
✅ Those who have experienced AI forgetting earlier context during long tasks
✅ Those who have watched several AI coding courses but ultimately had nothing tangible to show for it
Requirements
📌 Windows 10/11 or macOS PC (installation permissions required)
📌 platform.kimi.ai account + API key - issued together in Lesson 2
📌 A small top-up of API credits — the only payment required for this course
📌 Free Vercel account (sign up during the course for the deployment exercise in Session 8)
📌 Programming experience: Not required
※ Part 3 (Lessons 9–11) is advantageous if you have experience using Claude Code, but it is not required. Guidance is provided starting with installation.
Curriculum – 14 lessons, approximately 102 minutes
Part 1. Getting Started (approximately 19 minutes)
Lesson 1. Orientation - A Fable-Class Open Model Has Arrived (8 minutes)
Lesson 2. Setup — Obtaining an API key, installing Kimi Code, and a 5-minute setup (6 minutes)
Lesson 3. First Vibe Coding - Build an App with Just One Sentence (5 min)
Part 2. Kimi Code Vibe Coding in Practice (approximately 42 minutes)
Lesson 4. The Vibe Coding Loop - Ask, Check, and Have It Fix Things (9 min)
Lesson 5. Vision Vibe Coding - Give It a Single Design Mockup and Have It Build the UI (8 minutes)
Lesson 6. Mini Project ① - Have It Build the Entire Full-Stack App: From Planning to Implementation (12 min)
Lesson 7. Mini Project ② - Completion and Verification (6 minutes)
Lesson 8. Deployment - Releasing the App You Built to the World (7 minutes)
Part 3. Claude Code × K3 (Approx. 23 minutes)
Lesson 9. Direct API Setup - Integrating K3 into Claude Code (9 min)
Lesson 10. Getting K3 to Work in Claude Code - Hands-On (6 min)
Lesson 11. Head-to-Head — The Same Task, Kimi Code vs Claude Code (8 minutes)
※ Part 3’s setup is not officially supported by Anthropic and may change depending on client updates.
For details, please refer to the notice in item ⑤ above.
Part 4. The Road to Becoming a Power User (about 18 minutes)
Lesson 12. 1M Context - Complete Long Tasks from Start to Finish Without Summarizing (8 minutes)
Lesson 13. Pricing Breakdown — Is It Really the Price of a Cup of Coffee, and When Should You Use K3? (5 minutes)
Lesson 14. Limitations and Tips - An Honest Review and Next Steps (5 minutes)
14 lessons · approximately 102 minutes
※ The appendix, “The Same PRD, the Same Model, Different Tools,” will be released later. The video referred to as the “appendix” in Lesson 12 is this one, and it will be uploaded separately from the 14-lesson main course. It will not affect your ability to take the main course.
"How to Get AI to Do the Work" Series
This course was created using the same approach as the "How to Get ~ to Work for You" series.
Instead of memorizing tools, it helps you develop the feel for delegating tasks and verifying the results.
If you have never asked AI to do a task before, we recommend starting with the free course
「Ask AI to Work for You for the First Time - Claude Code Agentic Automation」.
This introductory course shows you how to ask it, in a single sentence each, to organize files, write documents, analyze data, automate tasks, and even create web pages.
14 years of R&D consulting · Over KRW 12.7 billion in cumulative government project funding · Over 3,700 cumulative students in Inflearn AI courses
Hello, I’m Park Hyun-il, co-CEO of FlowCoder.
For 14 years, I have provided consulting on R&D, new business planning, and government support programs for technology companies.
We have secured over KRW 12.7 billion in cumulative government projects and provided consulting to approximately 60 companies.
Since 2026, I have continued developing AI education, automation solutions, and SaaS with my colleague, Yonghyun Cho, after founding FlowCoder together.
Through corporate training, I have provided practical AI training at Samsung Electronics, Samsung Securities, Samsung Biologics, Hyundai Motor Company, Hyundai AutoEver, SK ecoplant, SK E&S, KB Investment, and other companies.
Two of my AI courses on Inflearn have been taken by a cumulative total of 3,700+ students and have a rating of 4.8.
I have been using agentic coding tools every day since March 2026. I switch between different tools.
While creating this course, I also built an app with K3, deployed it, and checked the invoice.
All the numbers shown here are actual costs I incurred.
"Tools keep changing. How to make requests and how to validate them remain."
Yes, the course is free. However, you’ll need Kimi API credits for the exercises, which you can purchase directly on platform.kimi.ai.
This is the only payment required by the course. You can start with a small top-up.
Q2. How much does the hands-on practice cost?
Based on the recording sessions, one session costs around 1–2 dollars ($1.41 on average across 7 sessions). A session in which an entire app was built and deployed, such as sessions 6–8, cost $1.69, while the average across 24 sessions, including additional experiments, was $5.10.
Those following along are likely to spend less than the instructor. Since the instructor was recording, the same exercises had to be run multiple times.
However, if you run it all day every day, the cost will not be this amount. It is pay-as-you-go. Session 13 also covers how to calculate the cost.
Q3. Can I follow along even if I know nothing about programming?
Yes. You won’t write any code yourself. You’ll write prompts, review the results, and explain what needs to be fixed.
How to write prompts (goals, scope, criteria, and constraints) is covered separately in Lesson 4.
Q4. Does it work on Windows too?
Yes. The installation process is covered for both Windows and macOS.
On Windows, installation is done using PowerShell commands, and Git for Windows must be installed first.
Q5. Can I take the course if I’ve never used Claude Code?
Yes. Sessions 1–8 (Parts 1 and 2) are conducted entirely using only Kimi Code CLI. This accounts for 60% of the total course content.
Part 3 covers Claude Code, but we provide guidance starting from installation, so you can follow along even if you’re a beginner.
Q6. Is connecting Kimi to Claude Code an officially supported method?
This is not an officially supported method by Anthropic. Anthropic's documentation states that routing Claude Code to non-Anthropic models is not supported, and its behavior may change when the client is updated. This course uses only a Kimi API key without Anthropic subscription credentials, and keeps the configuration within the project folder. Additionally, web reading does not work with this setup, and account integration tools are disabled. We disclose this as is and cover workarounds in Lesson 10.
Q7. I already have a Kimi account. Can I receive the 15% credit benefit?
No. This benefit only applies to newly registered accounts. It will not apply if you click the link using an existing account.
You can take the course without the benefit, and it will not affect the hands-on exercises at all.
Q8. Can I take the course with a membership?
Membership is currently operated through a waitlist application system.
This course is conducted from start to finish based on API pay-as-you-go pricing, so if you’re starting now, we recommend getting an API key.
Q9. How long is the course?
There are 14 lessons totaling approximately 102 minutes. You can watch them in separate Parts, and Parts 1 and 2 alone will take you through deploying an app.
Q10. Where can I ask questions during the course?
If you leave your question on the Inflearn Q&A board, I will answer it personally.
If the same questions accumulate, I will compile them into supplementary materials and upload them to the resources section.
Start now
The course is free, and there is no expiration date for access.
All you need is an API key and credits costing about the price of a cup of coffee.
It takes 102 minutes to create an app by telling it what to do and share it via a link.
Recommended for these people
Who is this course right for?
People who subscribe to AI coding tools but feel it’s a waste because they only use them two or three times a week.
Planners, marketers, and operations professionals with no programming experience who want to build and use the tools they need themselves
For those who are already using Claude Code and want to try a different model while keeping their familiar environment unchanged
Team lead considering a pay-as-you-go API-based alternative due to the cost structure
Need to know before starting?
No programming experience is required. AI writes the code, while you make requests and review it.
A PC running Windows 10/11 or macOS is required (installation privileges needed).
A platform.kimi.ai account and API key are required. We’ll go through the sign-up process together in Session 2.
A small top-up of API credits is needed. The entire course will be conducted using this one key, and this is the only payment.
In an era where AI writes code, we at FlowCoder design the 'Flow'
We are FlowCoder, a team that plans and develops services by designing the flow between humans and agents, and teaches that expertise through our lectures.
We design courses that enable even non-developers to build practical systems using AI agents.
Samsung · Hyundai · SK · KB · Hansol, etc. Claude · Vibe Coding · PBL practical training for executives and employees of major corporations
「Claude Code Development Automation System: Escaping Prompt Hell」 Author
「Claude Cowork Workbook」 scheduled for publication
![[Integrated] Vibe Coding for Beginners + Practical Web Build CompletionCourse Thumbnail](https://cdn.inflearn.com/public/files/courses/342990/cover/01kwkcejhccgf82zdekg0w61sb?w=420)
![[Integrated] Vibe Coding for Beginners + Complete Practical App DevelopmentCourse Thumbnail](https://cdn.inflearn.com/public/files/courses/342986/cover/01kwkcfpxw9bwfvp6018ap5w1n?w=420)
![[Practical AIoT] Perfect Preparation for Smart Mirror Makerthon: LLM, CV, and Hardware DesignCourse Thumbnail](https://cdn.inflearn.com/public/files/courses/340196/cover/01kexgfr26whtfsmsqd2dj1x7x?w=420)
![[AI Cheat Sheet] The secret to finishing work instantly, Agentic AICourse Thumbnail](https://cdn.inflearn.com/public/files/courses/340717/cover/ai/1/3b5cb844-25b5-4576-8224-d293d0989376.png?w=420)
![[Free] Notion MCP: From Beginner to ApplicationCourse Thumbnail](https://cdn.inflearn.com/public/files/courses/338906/cover/01k4nkjw13vgrmrs0esxnc60w9?w=420)