
How to use Qwen: step-by-step tutorial
Alibaba’s free AI chat and model family: setup, real tasks, the per-token API, privacy and how it compares, with prompts you can copy today.
At a glance
Is Qwen right for this job?
Use Qwen for
- Writing and translating between many languages
- Explaining or drafting code for a small project
- Summarizing long articles you paste in
- Brainstorming without paying for a plan
- Developers testing cheap, capable models over an API
Think twice for
- Legal or medical translations nobody fluent will check
- Live facts, unless you open the sources yourself
- Confidential company data on a personal account
- Jobs that need a guaranteed usage limit in writing
Get set up: step by step
Basics
What it is
Qwen is Alibaba’s family of AI models and the free chat app built on them. The official home page at qwen.ai describes Qwen Chat as “free to use, open to all”, with apps for phone and desktop. Developers reach the same models through Alibaba Cloud Model Studio, which charges per token.
Some third-party sites now call the chat app “Qwen Studio”. We could not confirm that name on an official Qwen page on Oct 30, 2026, so we use “Qwen Chat”, the name qwen.ai uses.
Why this step matters: Knowing what it is good and bad at stops you using it for jobs where it will fail quietly.
Setup
Step 1: Get the official app
- Start from qwen.ai and follow its links to the web chat or the phone and desktop apps.
- In an app store, check the publisher is Alibaba (or its Qwen team) before you install. Copycat “Qwen AI” apps exist.
- Sign in so your history is saved, then open the settings once to see what is stored.
Why this step matters: Copycat apps and unofficial installers are common; starting from the official source protects your account and data.
Setup
Step 2: Plans, limits and costs
Qwen Chat is free to use (qwen.ai). There is no paid consumer plan on the official page we checked.
The API is paid per token on Alibaba Cloud Model Studio. Example from the official price list: qwen3.8-max costs $2 per million input tokens and $6 per million output tokens. Prices differ by model and region, so read the table for the model you pick.
Why this step matters: Paying for features you don't use is the most common waste. Knowing the limits first tells you if the free tier is enough.
Real things to do with Qwen

Answer customers in two languages
Write the reply once, approve it, then ask for translations plus a list of phrases Qwen was unsure about. You check the risky bits instead of the whole thing.
Write a 90-word reply to this customer complaint. Friendly, one apology, one next step.
Then translate it into Spanish and French.
List any phrase you were unsure how to translate.
Understand code you inherited
Paste one function at a time. Ask for an explanation first and changes second, so you learn what it does before anything moves.
Explain this function line by line for a junior developer.
Then list 3 inputs that might break it, and a test for each.
Turn a long article into notes
Ask for bullets that each quote the source text. Quotes are fast to verify and stop invented points.
Summarize the article below in 6 bullets.
For each bullet, quote the sentence it comes from.
End with 2 questions the article leaves open.
Plan a lesson or workshop
Give the audience, time and goal. Ask for a timed plan and one activity per section.
Plan a 45-minute workshop on budgeting for first-year college students.
Timed sections, one activity each, and a 5-question exit quiz.
Draft a data-cleaning script
Describe the columns and the mess. Ask for a script plus a check that counts rows before and after.
Write a Python pandas script that cleans a CSV with columns date, amount, customer.
Dates are in mixed formats. Print row counts before and after, and list rows it dropped.
Test the API cheaply
If you build apps, run 10 real inputs through a small Qwen model on Model Studio and compare cost and quality against your current model before switching.
You are grading answers. For each pair (A, B) below, say which is more correct and why in one line.
Return JSON: [{"id":1,"winner":"A","reason":"..."}]The 5-part prompt that works in Qwen
Most weak answers come from weak instructions, not a weak model. Before you send anything important to Qwen, check that your message has these five parts. It takes thirty seconds and saves several rounds of “that’s not what I meant.”
- Situation. Who you are and why you are asking. “I run a two-person bakery” changes every answer that follows.
- Task. One clear verb: draft, compare, summarize, fix, plan, explain. If you have three tasks, send them as three steps.
- Inputs. Paste or attach the material it should use, and say what to ignore. Name the file and the pages if it is long.
- Format. Length, structure and tone: “a table with 4 columns”, “under 150 words”, “plain English, no jargon”.
- Checks. Tell it how you will judge the answer: “give a source link for every number”, “say unknown instead of guessing”, “list your assumptions first”.
Save the version that works. In Qwen the natural home for reusable instructions is one chat per project with a short brief at the top, so you never type the same context twice.
Go further: steps continued
Use
Step 3: Your first real task
Qwen is strong across many languages. A good first test is a real message you need in two languages:
I run a small online shop. Write a 90-word reply to a customer whose parcel is late. Friendly, one apology, one clear next step. Then give the same reply in Spanish and in Simplified Chinese. List any phrase you were unsure how to translate.Why this step matters: A prompt with situation, task and format gets a usable answer first time, so you learn what the tool can actually do.
Setup
Step 4: Privacy
- Don't paste passwords, ID numbers, card numbers or other people's private details.
- Assume the company can read and keep your chats unless its official policy says otherwise.
- For health, money or legal decisions, confirm with a professional or an official source.
- For work data, read Alibaba Cloud’s terms for Model Studio rather than the consumer app’s terms; they are different contracts.
Why this step matters: Anything you paste may be stored and read. Deciding what never goes in is easier than undoing a leak.
Mistakes
Mistakes
- Assuming “free” means unlimited. Heavy use can be slowed or limited; we could not find a published limit.
- Downloading an unofficial “Qwen” app from a lookalike publisher.
- Using the API without checking which model and region you are billed for.
- Taking facts without sources.
Why this step matters: Knowing how it typically fails tells you where to look when you review its work.
A good session, start to finish
Decide what “done” looks like
Write one sentence describing the finished result before you open the chat: “a one-page plan I can send to my landlord by Friday.” If you cannot describe it, the bot cannot hit it.
Give context once, in the right place
Put standing facts about you, your business or your project where {name} keeps them between chats, instead of pasting them into every message.
Ask for a draft, then steer
Take the first answer as a rough draft. Reply with specific edits (“shorter”, “cut the second option”, “use our real prices”) rather than starting over.
Check the parts that matter
Open the source behind every number, name, date and quote you will reuse. For files, spot-check two page references. For code, run it.
Save what worked
Copy the final prompt into a notes file or saved instructions. Next time the same job takes one message instead of ten, and you spend fewer credits doing it.
Your first week with Qwen
How to spot a wrong answer
Warning signs
- It gives an exact number, date or quote with no link, or with a link that does not contain it.
- It agrees with everything you say, including a wrong assumption you slipped in to test it.
- The answer is fluent but vague: lots of “generally” and “it depends” and no specifics you can act on.
- It claims to have done something (sent, saved, booked, run) without a link, ID or file you can open.
- Two answers to the same question, asked in fresh chats, disagree on a fact.
- It names a setting, menu or plan feature you cannot find. Apps change their menus often, so check the official help pages.
What to do
Open the source and find the exact sentence. Ask Qwen to separate what it checked from what it assumed. Re-ask in a fresh chat with the source attached. If it still cannot back the claim up, treat it as unknown and use the official page instead.
None of this means the tool is bad. It means you are the editor, and editors check.
Power-user tips
End important prompts with “list anything you are unsure about”. Qwen is good at flagging weak spots when asked.
Get the reply right in your language first, then translate. Fixing meaning in two languages at once doubles your work.
Long chats drift. Start a new chat with a 5-line brief when the subject changes.
For summaries, ask for a quoted sentence behind each point so you can Ctrl+F it.
Set a spending alert in Alibaba Cloud before your first API script runs in a loop.
If you self-host a Qwen model, read that model card’s licence before using it commercially.
Privacy and safety with Qwen
The consumer chat and the Alibaba Cloud API sit under different terms. Read the one that applies before you paste work data, and keep private data out of the free chat.
- Never paste passwords, API keys, one-time codes, full card numbers or government ID numbers into any chat.
- Remove other people’s names, addresses and account details from files before you upload them, unless you have their permission and a reason.
- Open the privacy or data-controls settings in your first week and decide whether your chats may be used to improve models.
- Use the account your employer approved for work material. A personal plan follows the consumer terms, not your company’s contract.
- For health, legal, tax or money decisions, use the bot to prepare questions, then confirm with a professional or an official source.
Qwen or another bot?
| If you need… | Pick | Why |
|---|---|---|
| Free chat in many languages | Qwen | Free on qwen.ai with strong multilingual output. |
| Cheap model API | Qwen or DeepSeek | Both price per token well below many Western rivals; test on your own inputs. |
| Current facts with links | Perplexity | Answers are built around citations. |
| Work inside Google apps | Gemini | It is built into Gmail, Docs and Drive. |
Our editorial judgment, not a benchmark. Try your own real task in two bots before you pay for either.
Questions people ask
Is Qwen free?
Qwen Chat is described on qwen.ai as “free to use, open to all” (checked Oct 30, 2026). The API on Alibaba Cloud Model Studio is paid per token.
Who makes Qwen?
Alibaba’s Qwen team. The models are offered through Alibaba Cloud Model Studio for developers.
Is it called Qwen Studio now?
Some third-party sites use “Qwen Studio”. qwen.ai still says Qwen Chat on Oct 30, 2026, so we can’t confirm the rename.
How much does the Qwen API cost?
It depends on the model. For example, the official Model Studio price list shows qwen3.8-max at $2 per million input tokens and $6 per million output tokens.
Is there an app?
Yes. qwen.ai links phone and desktop apps. Check the publisher before installing.
Can I use Qwen for coding?
Yes, for explaining and drafting code. Always run the code and tests yourself.
Is my data used for training?
We did not find a clear statement on the pages we checked. Assume chats may be stored and keep private data out.
Can I run Qwen myself?
Many Qwen models are released with open weights. Check each model card’s licence before commercial use.
Words you will see
The Qwen quiz, first-week checklist and cost calculator now have their own page.
Where these facts come from
- qwen.ai home: Qwen Chat “free to use, open to all”, phone and desktop apps
- Alibaba Cloud Model Studio pricing: per-token API prices (e.g. qwen3.8-max)
- The “Qwen Studio” name appears only on third-party sites (not confirmed officially)
Checked Oct 30, 2026. Plans change, so confirm on the official page before you pay.
Next: Qwen lessons →