These are 24 prompts worth saving for writing with Qwen3.7-Max, the flagship released in May 2026 with a 1M-token context window — enough room to paste your style guide, past drafts, and research and still get a long piece back. They are grouped by job: drafting from scratch, editing and tone, long-form structure, emails, social, and marketing copy. Run them free at chat.qwen.ai or in the iOS and Android apps.

Every prompt is written out in full and ready to copy. They follow the Role + Task + Context + Constraints + Output format shape, and the trick that makes Qwen sound like you is context: paste real samples of your writing, name the audience, and give the exact shape of the output. Leave Thinking off for quick drafting so copy comes back fast; turn it on only when the structure itself needs reasoning. Turn on Web Search when a piece needs a current fact or figure. New to Qwen? Start with the 40 best Qwen prompts roundup, and the Qwen prompt templates give you reusable skeletons. Swap anything in [BRACKETS] for your own details.

Advertisement

Drafting from scratch

The blank page is where rich context pays off most. Give Qwen the audience, the angle, and a sample of your voice, and it produces a real first draft instead of generic filler. Keep Thinking off (Fast or Auto) so drafts come back quickly.

1. Blog post from an outline

You are a writer for [AUDIENCE]. Write a [1,200]-word blog post titled "[TITLE]" from this outline: [PASTE OUTLINE]. Lead each section with its takeaway, use short paragraphs and concrete examples, and keep the tone practical and direct. Match this voice — study its sentence length and word choice: [PASTE 2-3 SAMPLES OF MY WRITING]. Avoid clichés and filler. Output clean Markdown with H2 sections and a two-line intro.

Best for: Turning a rough outline into a publishable draft that reads like you wrote it, not a template.

2. Full article from a working title

Write a [900]-word article for [PUBLICATION / BLOG] aimed at [AUDIENCE]. Working title: "[TITLE]". Angle: [THE ONE POINT IT SHOULD MAKE]. Cover these must-hits: [LIST KEY POINTS]. Open with a specific, concrete hook — no throat-clearing — then build the argument with examples, and close with a takeaway the reader can act on. Keep the tone [DESCRIBE]. Output in Markdown with H2s and a one-line meta description at the top.

Why it works: Naming the single point and the must-hit list keeps the article focused instead of wandering across the whole topic.

3. Fast first draft from messy notes

Here are my raw notes for a piece on [TOPIC]: [PASTE NOTES]. Turn them into a coherent first draft of about [LENGTH] for [AUDIENCE]. Keep every idea and fact in my notes, organize them into a logical flow, fill the gaps between points with connective prose, and flag anything that's thin or contradictory rather than papering over it. Don't add claims I didn't make. Plain, direct tone.

Best for: Getting from a pile of bullet points to a draft you can edit, without losing anything you jotted down.

4. Brainstorm fresh angles on a topic

I'm writing about [TOPIC] for [AUDIENCE]. Give me 10 angles that aren't the obvious ones everyone else covers. For each: a one-line angle, who it would most resonate with, and the single strongest example or data point that would support it. Mix contrarian takes, a personal-story angle, a "what everyone gets wrong" angle, and a practical how-to. Then pick the three with the most potential and say why.

Why it works: Asking for the strongest supporting point per angle filters out ideas that sound clever but have nothing behind them.

Editing & tone

Qwen is a sharp editor when you tell it exactly what to change and what to leave alone. Because it is strongly multilingual, it is also excellent at polishing non-native English into natural, fluent prose without flattening your meaning. Keep Thinking off for a quick pass; turn it on for a deep structural rewrite.

5. Rewrite for clarity

Rewrite the text below to be clearer and more direct without losing meaning. Cut filler, fix weak verbs, break up long sentences, and keep my key terms and my point exactly. Target a [confident but plain] tone at a [general professional] reading level. Return the rewrite first, then a short bullet list of the main changes you made and why. Text: [PASTE TEXT]

Best for: Copy that says the right thing but reads like a slog. The change list lets you keep control.

6. Tighten without losing meaning

Cut this down to [TARGET WORD COUNT] words while keeping every essential point and my voice: [PASTE TEXT]. Remove redundancy, hedging, and throat-clearing first; only cut real content if you have to, and if you do, tell me what went. Don't make it choppy — it should still read smoothly. Return the tightened version, then note the word count before and after.

Why it works: Telling Qwen to cut filler before content means the tightening comes from waste, not from the ideas you actually need.

7. Shift the tone

Rewrite this in a [WARMER AND MORE CONVERSATIONAL] tone for [AUDIENCE], keeping the meaning and all the facts the same: [PASTE TEXT]. Adjust word choice, sentence rhythm, and formality to fit that tone, but don't add fluff or exclamation-mark energy it doesn't need. Show me the rewrite, then a one-line note on what you changed to move the tone.

Best for: Taking the same message from stiff to friendly (or the reverse) for a different reader or channel.

8. Proofread and polish non-native English

English is not my first language. Proofread and polish this text so it reads like fluent, natural English written by a native speaker, without changing my meaning or making it sound stiff or overly formal: [PASTE TEXT]. Fix grammar, articles, prepositions, and awkward phrasing, and smooth anything that sounds translated. Return the corrected version, then list the recurring mistakes you noticed so I can learn them.

Why it works: Qwen's multilingual strength makes it unusually good at hearing what "translated" English sounds like, and the recurring-mistakes list turns one edit into a lesson.

Long-form & structure

This is where Qwen3.7-Max's 1M-token context shines — paste all your research, interviews, and source docs at once and let Qwen hold the whole thing while it structures a long piece. Turn Thinking on here so it plans the argument before it writes. For document-heavy work, see the Qwen prompt templates.

9. Structure a long report

Turn on Thinking. I'm writing a report on [TOPIC] for [AUDIENCE]. Here is all my research and raw material: [PASTE OR ATTACH EVERYTHING]. Before drafting, propose a structure: the main sections in order, what each one argues, which pieces of my material support it, and where the gaps are that I still need to fill. Flag the strongest and weakest parts of the evidence. Don't write the report yet — just the structure and the gap list.

Why it works: Pasting everything into the 1M context lets Qwen map your actual material to a structure, so the outline is grounded in what you have, not a generic template.

10. Detailed outline from research

Turn on Thinking. Build a detailed outline for a [LENGTH] piece on [TOPIC] for [AUDIENCE], based on this research: [PASTE OR ATTACH]. For each section, give the heading, the key point it makes, two or three supporting bullets drawn from the research, and a rough word budget. Make sure the sections build a logical argument from start to finish. Note anywhere the research is thin so I know where to dig more.

Best for: A working blueprint you can hand back to Qwen section by section, or write yourself.

11. Draft a whitepaper section

Turn on Thinking. Draft the "[SECTION NAME]" section of a whitepaper on [TOPIC] for [TECHNICAL / EXECUTIVE] readers. Context, so it stays consistent with the rest: [PASTE OUTLINE + KEY FACTS + ANY EARLIER SECTIONS]. Target [LENGTH]. Be precise and evidence-led, define terms on first use, and use subheadings and a short summary box at the end. Don't overstate — mark any claim that would need a citation with [cite].

Why it works: Feeding earlier sections into the context keeps terminology and claims consistent, and the [cite] markers show you exactly what still needs a source.

12. Turn a transcript into a written piece

Attached is a transcript of [A TALK / INTERVIEW / MEETING]. Turn it into a well-structured [ARTICLE / SUMMARY / Q&A] of about [LENGTH] for [AUDIENCE]. Keep the speaker's ideas and best quotes, but fix the rambling, cut the filler, and give it a clear arc with headings. Don't invent anything that isn't in the transcript. Preserve any exact quotes you use verbatim and mark them as quotes.

Best for: Repurposing spoken content into something readable without losing the original voice or facts.

Advertisement

Emails & messages

The fix for AI-sounding email is context: tell Qwen who you are, who you're writing to, and the real goal. Paste a couple of your past emails so it catches your voice. For anything sensitive, ask for two versions and pick the one that fits.

13. Cold email in your voice

Write a cold outreach email from me ([MY ROLE] at [MY COMPANY]) to [PROSPECT ROLE] at [PROSPECT COMPANY]. Goal: [GOAL]. Keep it under 90 words, open with a specific reason I'm reaching out to them (not a template line), state one clear value point, and end with a low-friction ask. Match my voice from these past emails: [PASTE 1-2 SAMPLES]. Give me three subject-line options, none clickbaity.

Best for: Outreach that reads like a person wrote it, with subject lines you can A/B test.

14. Follow-up email that isn't pushy

Write a follow-up email to [WHO] about [WHAT], since I haven't heard back after [TIME]. Here's the last message and any context: [PASTE THREAD]. Keep it short and warm, add one new reason to reply (a useful detail, not just "checking in"), make it easy to say yes or no, and don't guilt them. Give me a two-line version and a slightly fuller version.

Why it works: Adding a genuine new reason to reply is what separates a useful follow-up from an annoying nudge.

15. Reply drafted in your voice

Draft a reply to this email in my voice: [PASTE THE EMAIL]. What I want to get across: [YOUR POINTS / DECISION / ASK]. Constraints: [e.g. stay friendly, hold my position, don't over-commit]. Match my style from these examples: [PASTE 1-2 OF MY EMAILS]. Keep it as short as it can be while covering everything. Flag anything in their email I should address but didn't mention.

Best for: Clearing a reply out of your inbox fast, in your tone, without missing what they actually asked.

16. Difficult message, handled well

Help me write a message to [WHO] about [SENSITIVE SITUATION]. My goal is [GOAL] and I want to sound [TONE — e.g. firm but kind]. Keep it honest and clear, don't over-apologize or ramble, own what's mine to own, and leave the door open for a good response. Context they'll need: [BACKGROUND]. Give me two versions — a more direct one and a warmer one — and note the risk of each.

Why it works: Two versions with their trade-offs spelled out let you choose the tone that fits the relationship instead of guessing.

Social & short-form

Short-form lives or dies on the first line. Give Qwen the platform, the audience, and the one idea, and ask for several options so you can pick the strongest hook. Turn on Web Search if a post needs a current stat or reference.

17. LinkedIn post that earns attention

Write a LinkedIn post for [MY ROLE / AUDIENCE] about [IDEA OR STORY]. Open with a strong first line that stands alone in the feed, keep paragraphs to one or two lines with white space, make one clear point, and end with a question or takeaway that invites comments. No hashtag spam — three relevant tags max. Keep it human, not corporate. Give me two versions with different opening hooks.

Best for: A post that reads naturally and gives people a reason to stop scrolling and reply.

18. X thread from one idea

Turn this idea into an X thread of [7-10] posts: [PASTE IDEA OR ARTICLE]. Post 1 is the hook — make it specific and impossible to scroll past, no "a thread 🧵" throat-clearing. Each following post makes one point and stands on its own, under 280 characters. Build momentum, put the best insight in the middle, and end with a payoff or clear CTA. Keep my voice plain and confident. Number the posts.

Why it works: Forcing each post to stand alone and capping the character count produces a thread that actually works on the platform, not a chopped-up essay.

19. Captions for a batch of posts

Write captions for [NUMBER] [INSTAGRAM / social] posts for [BRAND / ME]. Here's what each post is about: [LIST]. For each, give a scroll-stopping first line, two or three lines of caption in a [TONE] voice, a soft CTA, and five relevant hashtags. Keep them varied — don't reuse the same opener structure. Match this brand voice: [DESCRIBE OR PASTE SAMPLES].

Best for: Batching a week of captions in one pass while keeping each one distinct.

20. Hook variations for one piece

Give me 12 opening hooks for [PIECE / POST / VIDEO] aimed at [AUDIENCE]. The core idea is [IDEA]. Mix angles: a bold claim, a surprising stat, a relatable pain point, a short story opener, a question, and a "most people get this wrong" line. Keep each under [20] words, no hype words, and label the angle of each. Then pick your top three and say why they'd stop the scroll.

Why it works: A batch of labeled hooks with the reasoning attached gives you testable openers instead of one take-it-or-leave-it line.

Marketing & copy

Marketing copy needs the offer, the audience, and the real benefit up front — vague inputs give vague copy. Give Qwen the product details and who it's for, and demand specifics over adjectives. See the full set in the best Qwen prompts roundup.

21. Landing page copy

Write landing page copy for [PRODUCT], which helps [AUDIENCE] to [OUTCOME]. Key details: [FEATURES, BENEFITS, PROOF POINTS, PRICE]. Give me: a headline and subhead, three benefit blocks (each a benefit-led heading plus two lines), a short "how it works" in three steps, an objection-handling FAQ of three items, and one primary CTA. Lead with benefits, not features; be concrete; no hype words. Match this brand voice: [DESCRIBE OR PASTE].

Best for: A full page structure you can drop into a builder and refine, not just a headline.

22. Ad copy variations

Write [8] ad variations for [PRODUCT] targeting [AUDIENCE] on [PLATFORM]. The main benefit is [BENEFIT] and the offer is [OFFER]. Vary the angle — pain point, benefit, social proof, curiosity, urgency — and label each. Respect [PLATFORM]'s length limits, put the hook in the first few words, and end each with a clear CTA. No hype adjectives; make the value concrete. Then say which two you'd test first and why.

Why it works: Labeled angles within the platform's real limits give you a test matrix, not eight near-identical lines.

23. Product description that sells

Write a product description for [PRODUCT]. Details: [FEATURES, MATERIALS, SPECS, WHO IT'S FOR]. Structure it as a one-line hook, a short paragraph that sells the outcome and feeling, then a scannable bullet list of key features each tied to a benefit. Keep it [TONE], honest, and specific — no empty superlatives. Target about [LENGTH]. Add a short meta description for the product page.

Best for: E-commerce copy that connects each feature to why a buyer should care.

24. Sharpen a value proposition

Help me sharpen the value proposition for [PRODUCT / COMPANY]. Here's what it does, who it's for, and how it's different: [PASTE DETAILS]. Draft five one-sentence value propositions, each clearly stating who it's for, the outcome it delivers, and why it beats the alternative. Vary the emphasis across them. Then critique each in one line — what's strong, what's vague — and recommend the sharpest one.

Why it works: Making Qwen critique its own drafts surfaces the vague, everyone-says-this lines so you keep only the one that actually differentiates you.

Writing in another language, or need to adapt a piece for a new market? The Qwen translation prompts lean on Qwen's multilingual strength, and the best Qwen prompts roundup collects the full set across every job.

Frequently Asked Questions

Which Qwen model are these writing prompts built for?

They target Qwen3.7-Max, the flagship released in May 2026, with a 1M-token context window so you can paste a whole style guide, past drafts, and research and still have room for a long piece. They also work on Qwen-Plus and the free Qwen Chat tiers, since Thinking mode, Web Search, and file upload are shared features.

How do I make Qwen match my voice?

Show it, don't describe it. Paste two or three samples of your real writing and tell Qwen to study the sentence length, rhythm, vocabulary, and how formal you are, then write in that voice. Qwen's 1M context means you can paste a lot of samples. Give it the role, the audience, and the exact output format too, and the draft comes back sounding like you.

Should I turn Thinking on or off for writing?

Leave Thinking off (use Fast or Auto) for straightforward drafting and editing so the copy comes back quickly and doesn't over-explain. Turn Thinking on when the structure itself needs reasoning — a long report outline, an argument that has to hold together, or a rewrite where you want Qwen to plan before it writes.

Can Qwen write in other languages?

Yes. Multilingual is one of Qwen's strengths — it writes natively in 100+ languages and is especially strong across Chinese, English, and Asian languages. Ask it to draft in any language, or draft in one and adapt to another. For dedicated translation work, see the Qwen translation prompts.

Is Qwen free?

Yes. Qwen Chat is free at chat.qwen.ai and in the iOS and Android apps, and the open-weight Qwen3 models are on Hugging Face and GitHub under the Apache 2.0 license, so you can self-host. The flagship Qwen3.7-Max is available through paid tiers, but the free chat handles the writing prompts here well.

How do the [BRACKET] placeholders work?

Anything in square brackets is a swap-in. Replace [TOPIC], [PASTE DRAFT], or [AUDIENCE] with your own details before sending. The rest of each prompt is written to be copy-paste ready.

Advertisement