Skip to content

Free tool · Articles

AI article summarizer

Paste an article and get its gist, key points and why it matters in 1 free run a day on GPT-6 Luna, no account needed. Every figure and name is kept as the article gives it, and claims stay attributed.

1 free run a day, no account needed

GPT-6 Luna

Up to 1,600 characters. No files, tools or web search in a free run. One free run a browser, and one a network each day, shared with the free message on our home page.

Try an example:

By running it, you agree to the Terms and Privacy Policy and confirm you’re 18 or older. We keep what you send and its answer for 7 days, or in your account if you sign up.

Test runs checked . Updated .

AI article summarizer: three examples

Three inputs and the summary a free run gave each, made once on September 28, 2026 with GPT-6 Luna and shown as it came back. The example buttons in the box above show the same.

A news story

The city council voted 7–2 on Tuesday to ban cars from the Old Market district on weekends, starting 1 March. The ban covers 14 streets and will be reviewed after six months. Mayor Elena Ruiz said the change would 'give the square back to people', and pointed to a 2025 trial in which weekend footfall rose 22%. Shop owners are divided: the Old Market Traders' Association said 40% of its members oppose the plan, fearing lost deliveries. Delivery vans will be allowed until 10 am, and residents' permits remain valid. Critics on the council said parking in nearby streets, already 90% full on Saturdays, would get worse.

Gist: The city council voted 7–2 to ban cars from the Old Market district on weekends starting 1 March.

- The ban covers 14 streets and will be reviewed after six months.
- Mayor Elena Ruiz said the change would “give the square back to people” and cited a 2025 trial in which weekend footfall rose 22%.
- The Old Market Traders' Association said 40% of its members oppose the plan, fearing lost deliveries; shop owners are divided.
- Delivery vans will be allowed until 10 am, and residents' permits will remain valid.
- Critics on the council said nearby streets, already 90% full on Saturdays, would face worse parking.

Why it matters: The article says the ban is intended to return the square to people, while opponents worry about deliveries and parking.
A research write-up

A new study of 2,300 office workers suggests that four-day weeks do not reduce output, at least in the short term. Researchers at the University of Leeds compared teams that switched to a 32-hour week with teams that did not, over nine months. Output per team, measured by tasks closed, stayed within 2% of the control group, while reported burnout fell by a third. The authors caution that most participants worked in software and marketing, and that the results may not hold in customer-facing jobs. Three of the 41 companies returned to five days before the study ended.

Gist: A study of 2,300 office workers found that a 32-hour, four-day week maintained short-term output while reducing reported burnout.

- Researchers at the University of Leeds compared teams that switched to a 32-hour week with teams that did not over nine months; output per team stayed within 2% of the control group.
- Reported burnout fell by a third, but the authors caution that most participants worked in software and marketing and results may not hold in customer-facing jobs.
- Three of the 41 companies returned to five days before the study ended.

Why it matters: The findings suggest four-day weeks may preserve output in the short term, though the authors say more types of work may not see the same results.
An opinion piece

Every few years a new tool promises to end the meeting. Chat did not; video made it worse. The problem was never the medium but the default: we meet because a calendar slot is free, not because a decision needs making. Teams that cut meetings successfully do three things. They write a one-page brief before any meeting that decides something. They cancel any recurring meeting nobody would recreate from scratch. And they protect two afternoons a week with no meetings at all. None of this requires software, which is perhaps why so few companies try it.

Gist: The article argues that reducing unnecessary meetings depends on changing meeting habits, not adopting new communication tools.

- Chat did not end meetings, and video made them worse; the article says the problem is meeting by default rather than because a decision needs to be made.
- Teams that successfully reduce meetings write a one-page brief before any meeting that decides something.
- They cancel recurring meetings nobody would recreate from scratch.
- They protect two afternoons a week with no meetings.
- The article says these changes require no software, which may help explain why so few companies try them.

Why it matters: The article argues that simple changes to how teams schedule and prepare for meetings can reduce unnecessary meetings without new software.

Which AI model is best at this? We tested every one

Every one of the 19 models we ran passed all 5 of our AI article summarizer test inputs, so choose on speed and price: GLM 5.3 was quickest, at 0.9 s a run on average. GPT-6 Luna, the model a free run is on, took 2.9 s.

Run on October 9, 2026: 5 inputs on 19 models, each with this tool's own instructions.

Each model's results on the tool's test inputs
ModelPassedHard onesMessages used on ProCost per runTime per run
GPT-6 LunaOpenAIThe free run's model5 of 52 of 21 each, of 60 a day on Pro$0.00012.9 s
GLM 5.3 FlashZ.ai5 of 52 of 21 each, of 60 a day on Pro$0.00012.2 s
Claude Haiku 5.5Anthropic5 of 52 of 21 each, of 60 a day on Pro$0.00021.6 s
DeepSeek V4.1 FlashDeepSeek5 of 52 of 21 each, of 60 a day on Pro$0.00041.6 s
Gemini 3.8 FlashGoogle5 of 52 of 21 each, of 250 a month on Pro$0.00051.4 s
GLM 5.3Z.ai5 of 52 of 21 each, of 250 a month on Pro$0.00090.9 s
Claude Haiku 4.5Anthropic5 of 52 of 21 each, of 250 a month on Pro$0.00121.9 s
DeepSeek V4 ProDeepSeek5 of 52 of 21 each, of 250 a month on Pro$0.00345.9 s
Grok 4.7xAI5 of 52 of 21 each, of 250 a month on Pro$0.00579.9 s
Mistral Large 4Mistral5 of 52 of 21 each, of 250 a month on Pro$0.009644.4 s
GPT-6.1 SolOpenAI5 of 52 of 21 each, of 125 a month on Pro$0.00102.4 s
GPT-6 SolOpenAI5 of 52 of 21 each, of 125 a month on Pro$0.00136.0 s
Kimi K3Moonshot5 of 52 of 21 each, of 125 a month on Pro$0.00187.1 s
Claude Sonnet 5.5Anthropic5 of 52 of 21 each, of 125 a month on Pro$0.00322.0 s
Claude Sonnet 5Anthropic5 of 52 of 21 each, of 125 a month on Pro$0.00333.0 s
Gemini 3.1 Pro (preview)Google5 of 52 of 21 each, of 125 a month on Pro$0.00709.6 s
Claude Opus 5.5Anthropic5 of 52 of 21 each, of 62 a month on Pro$0.00834.1 s
GPT-6 AstraOpenAI5 of 52 of 21 each, of 31 a month on Pro$0.00528.9 s
Claude Fable 5.1Anthropic5 of 52 of 21 each, of 31 a month on Pro$0.01673.4 s
Passed: of the 5 test inputs, how many answers passed their check (an answer the provider failed to give isn't counted). Messages used: what each run counted as on Pro, and Pro's count on the model. Cost: what OpenRouter charged us per run, on average; in llmwise you pay per message, not per token. Time: from sending to the whole answer.

Side by side: a recall with conditions

The same input on GPT-6 Luna and GLM 5.3 Flash, as each answered it, unedited.

Summarize this article: a "Gist:" line, the key points with every number and name exact, and a "Why it matters:" line.

---
Brightline is recalling some of its electric kettles after reports of faulty lids. The recall covers only kettles with serial numbers starting K7, sold between March and June 2026. Owners should stop using them and can choose a full refund or a free replacement by calling the number on the company's website. Kettles bought after 1 July are not affected, and no injuries have been reported.
GPT-6 Luna
Gist: Brightline is recalling some electric kettles because of reports of faulty lids.

- The recall covers kettles with serial numbers starting K7, sold between March and June 2026.
- Owners should stop using the kettles and can choose a full refund or a free replacement by calling the number on the company's website.
- Kettles bought after 1 July are not affected, and no injuries have been reported.

Why it matters: Owners of affected kettles can stop using them and seek a refund or replacement.
GLM 5.3 Flash
Gist: Brightline is recalling certain electric kettles over reports of faulty lids, offering owners a refund or replacement.

- The recall covers only kettles with serial numbers starting K7, sold between March and June 2026.
- Owners should stop using the kettles and can choose a full refund or a free replacement by calling the number on the company's website.
- Kettles bought after 1 July are not affected.
- No injuries have been reported.

Why it matters: If you own a Brightline kettle with a K7 serial number from that period, you should stop using it and contact the company for a refund or replacement.

How the inputs are scored

  • Key facts: the answer must state each fact the input turns on (a number, a name, a date or a condition), as the input gives it, keep any format rule the input sets (one sentence, a word limit, a number of items, a length for each), and say nothing the input rules out (an ingredient, a claim it doesn't back); how it's worded isn't judged.

The inputs: a fare rise, with its figures, a survey, with its source, an opinion, kept as one, a recall with conditions, and a short summary, three points at most. Every model gets the tool's own instructions and up to Pro's reply limit. The method is the same as our other test runs.

How to use the AI article summarizer

  1. Copy the article's text (not the link) and paste it in, up to 1,600 characters in a free run.

  2. Pick the length and the language.

  3. Press “Summarize”. Without an account you see the gist and the first points (1 free run a day, no account needed).

  4. Signed in, paste a longer read and ask follow-up questions about it.

Tips for a better answer

  • Paste only the article

    Copy the article's text, leaving out menus, ads and comments. Everything you paste is treated as part of the article.

  • Ask for an angle

    Add a line such as “focus on what it means for renters” and the key points lead with that angle.

  • Opinion stays opinion

    For a piece arguing one side, the summary keeps the argument as the author's. If you want the counter-case, ask for it in the chat afterwards.

  • Spot-check two numbers

    Before you share a summary, check two figures against the article. It takes ten seconds and catches the rare slip.

What a good summary looks like

  • Numbers as written

    “The central bank raised rates to 5.5%” should stay 5.5%, not “raised rates sharply”. A summary that rounds or drops the numbers can't be quoted.

  • Claims stay attributed

    If the article says “according to the company”, the summary should too. Turning a claim into a fact is the most common way summaries mislead.

  • The gist first

    The first line should tell you what happened or what the piece argues, so you can decide in five seconds whether to read the rest.

  • Nothing added

    Nothing the article doesn't say: no background from elsewhere, no opinion on who's right.

What's free, and what signing up adds

  • Without an account: 1 free run a day, on GPT-6 Luna, shared with the free message on our home page. It takes up to 1,600 characters, writes up to about 600 tokens (roughly 450 words), and shows you the start of the summary.

  • Signing up is free, with an email link and no card. llmwise accounts are for people 18 and older. The summary you ran comes with you as your first chat, and you get 5 free messages on almost every model (all but Claude Fable 5.1 and GPT-6 Astra).

  • Signed in, paste up to 60,000 characters in a run. It runs on the model you pick, on your plan, shows the whole answer, and you can carry on in the chat with any model.

More free tools are on the tools page, and what each plan gives is on pricing.

Questions

Is the article summarizer free?

Yes. Without an account you get 1 run a day on GPT-6 Luna, shared with the free message on our home page: you see the start of the summary, and the rest when you sign up. Signing up is free and adds 5 free messages on almost every model in llmwise (all but Claude Fable 5.1 and GPT-6 Astra; one of them can be on Claude Opus 5.5), with no card.

Can I paste a link instead of the text?

It reads the text you paste, not links: paywalls and cookie banners get in the way of fetching pages. Open the article, select the text and copy it in.

How long an article can it summarize?

A free run takes up to 1,600 characters, a short article. Signed in, paste a long read of 60,000 characters and ask about it afterwards in the chat.

How is this different from the other summarizers?

For a single sentence, use the one-sentence summary tool; for a PDF report, the PDF summarizer. This one is built for news, blog posts and essays: the gist, the points, and why it matters.

Which AI summarizes articles best?

Our test runs below gave every model the same five articles and checked that each summary kept the figures, names and conditions that matter.

Which AI is best for this, in your tests?

Every one of the 19 models we ran passed all 5 of our AI article summarizer test inputs, so choose on speed and price: GLM 5.3 was quickest, at 0.9 s a run on average. GPT-6 Luna, the model a free run is on, took 2.9 s. The table on this page has every model's result, what a run counts as on Pro, and the time it took.

Claude, GPT, Gemini, DeepSeek, Grok, Kimi, GLM, and Mistral, in one chat.

See what a message costs before you send it. Free is 5 messages to try; sign in with an email link, no password or card.