Free tool · YouTube
YouTube video summarizer
Paste a YouTube video's transcript and get its key points in under a minute: 1 free run a day on GPT-6 Luna, no account needed. It reads the text you paste; it doesn't open links or watch the video.
1 free run a day, no account needed
GPT-6 LunaBy 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 .
YouTube 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.
0:00 Today we're making a one-pan lemon garlic chicken that takes about thirty minutes. 0:12 You'll need four chicken thighs, one lemon, six cloves of garlic, a cup of chicken stock and some fresh thyme. 0:31 Season the thighs with salt and pepper and sear them skin-side down for about seven minutes, until the skin is really golden. Don't move them. 1:05 Flip them, take them out, and in the same pan soften the garlic for a minute. Keep the heat medium so it doesn't burn. 1:28 Pour in the stock and the juice of the lemon, scrape up the brown bits, and add the thyme. 1:46 Put the chicken back skin-side up and bake at 200 degrees Celsius for fifteen minutes. 2:10 Rest it for five minutes, spoon the sauce over, and serve it with rice or crusty bread.
This video shows home cooks how to make a one-pan lemon garlic chicken in about thirty minutes. 0:00 - The recipe takes about thirty minutes. 0:12 - You’ll need four chicken thighs, one lemon, six cloves of garlic, a cup of chicken stock and some fresh thyme. 0:31 - Season the thighs with salt and pepper, then sear them skin-side down for about seven minutes until golden; don’t move them. 1:05 - Flip the thighs and take them out. Soften the garlic in the same pan for a minute over medium heat so it doesn’t burn. 1:28 - Add the stock and lemon juice, scrape up the brown bits and add the thyme. 1:46 - Return the chicken skin-side up and bake at 200 degrees Celsius for fifteen minutes. 2:10 - Rest for five minutes, spoon the sauce over the chicken and serve with rice or crusty bread. Worth watching for: It adds a step-by-step cooking demonstration.
Most students study by rereading their notes, and it feels productive, but the research says it's one of the least effective methods. What works better is retrieval practice: closing the book and trying to recall what you just read. Even getting it wrong helps, because the effort of recalling is what strengthens the memory. The second technique is spacing. Instead of studying a topic for three hours the night before an exam, study it for thirty minutes on six different days. You'll forget some of it in between, and that's the point: relearning what you forgot makes it stick. The third one is interleaving, which means mixing different types of problems in one session instead of doing twenty of the same kind in a row. It feels harder, and your practice scores go down, but your exam scores go up. So the plan is simple: test yourself, spread it out, and mix it up.
This video explains three evidence-based study techniques for students: retrieval practice, spacing, and interleaving. - Rereading notes feels productive, but the video says research finds it is one of the least effective study methods. - Retrieval practice means closing the book and trying to recall what you read. Even getting it wrong can help because the effort of recalling strengthens memory. - Spacing means studying a topic for thirty minutes on six different days instead of for three hours the night before an exam. Forgetting some material between sessions is part of the process: relearning it helps it stick. - Interleaving means mixing different types of problems in one session instead of doing twenty of the same kind in a row. It may lower practice scores, but the video says exam scores go up. - The plan is: test yourself, spread it out, and mix it up. Worth watching for: The video adds no demonstration, chart, or story that the summary misses.
If you're starting to invest in India with a small amount, here's what I'd do. First, build an emergency fund of six months of expenses in a savings account or a liquid fund before you invest anything. Second, start a SIP in a low-cost index fund that tracks the Nifty 50, even if it's just five hundred rupees a month; the habit matters more than the amount at the start. Third, check the expense ratio: for an index fund it should be well under half a percent. Fourth, don't stop your SIP when the market falls, because that's when your money buys more units. And finally, review once a year, not every day. Checking your portfolio daily is the fastest way to make an emotional decision.
This video offers a simple starting plan for people who want to invest small amounts in India. First, build an emergency fund covering six months of expenses in a savings account or liquid fund; then start a SIP in a low-cost Nifty 50 index fund, even with just five hundred rupees a month, since building the habit matters more than the initial amount. Check that the index fund’s expense ratio is well under half a percent, keep investing when the market falls because your money buys more units, and review your portfolio once a year rather than daily to avoid emotional decisions. Worth watching for: it adds nothing the summary misses.
Which AI model is best at this? We tested every one
Every one of the 15 models we ran passed all 5 of our YouTube summarizer test inputs, so choose on speed and price: GLM 5.3 was quickest, at 1.3 s a run on average. GPT-6 Luna, the model a free run is on, took 6.6 s.
Run on September 28, 2026: 5 inputs on 15 models, each with this tool's own instructions.
| Model | Passed | Hard ones | Messages used on Pro | Cost per run | Time per run |
|---|---|---|---|---|---|
| GPT-6 LunaOpenAIThe free run's model | 5 of 5 | 2 of 2 | 1 each, of 60 a day on Pro | $0.0001 | 6.6 s |
| GLM 5.3 FlashZ.ai | 5 of 5 | 2 of 2 | 1 each, of 60 a day on Pro | $0.0002 | 2.5 s |
| DeepSeek V4.1 FlashDeepSeek | 5 of 5 | 2 of 2 | 1 each, of 60 a day on Pro | $0.0006 | 1.9 s |
| Gemini 3.8 FlashGoogle | 5 of 5 | 2 of 2 | 1 each, of 250 a month on Pro | $0.0006 | 2.0 s |
| GLM 5.3Z.ai | 5 of 5 | 2 of 2 | 1 each, of 250 a month on Pro | $0.0014 | 1.3 s |
| Claude Haiku 4.5Anthropic | 5 of 5 | 2 of 2 | 1 each, of 250 a month on Pro | $0.0018 | 2.7 s |
| DeepSeek V4 ProDeepSeek | 5 of 5 | 2 of 2 | 1 each, of 250 a month on Pro | $0.0047 | 7.4 s |
| Grok 4.7xAI | 5 of 5 | 2 of 2 | 1 each, of 250 a month on Pro | $0.0075 | 14.5 s |
| GPT-6 SolOpenAI | 5 of 5 | 2 of 2 | 1 each, of 125 a month on Pro | $0.0019 | 6.4 s |
| Kimi K3Moonshot | 5 of 5 | 2 of 2 | 1 each, of 125 a month on Pro | $0.0029 | 32.2 s |
| Claude Sonnet 5Anthropic | 5 of 5 | 2 of 2 | 1 each, of 125 a month on Pro | $0.0052 | 4.3 s |
| Gemini 3.1 Pro (preview)Google | 5 of 5 | 2 of 2 | 1 each, of 125 a month on Pro | $0.0084 | 10.9 s |
| Claude Opus 5.5Anthropic | 5 of 5 | 2 of 2 | 1 each, of 62 a month on Pro | $0.0137 | 5.3 s |
| GPT-6 AstraOpenAI | 5 of 5 | 2 of 2 | 1 each, of 31 a month on Pro | $0.0080 | 3.2 s |
| Claude Fable 5.1Anthropic | 5 of 5 | 2 of 2 | 1 each, of 31 a month on Pro | $0.0281 | 4.9 s |
Side by side: a headphone review, as a timestamped outline
The same input on GPT-6 Luna and GLM 5.3 Flash, as each answered it, unedited.
Summarize this YouTube video from its transcript. Start with one sentence on what it's about, then the key points in order. Make it a timestamped outline: each point starts with its timestamp from the transcript. --- [00:00] Hi, I'm Maya, and this is my review of the Aurora X2 wireless headphones after three weeks of daily use. [00:48] Build first. They weigh 254 grams, the headband is aluminium, and the ear cushions are replaceable, which is rare at this price. [01:42] Battery life: I measured 31 hours with noise cancelling on, against the 35 hours on the box. Ten minutes of charging gets you about five hours. [03:05] Noise cancelling is very good on planes and trains, but it struggles with voices in an open office. [04:10] Call quality is the weak point. People told me I sounded distant, even in a quiet room. [05:30] The app lets you adjust the EQ and pair with two devices at once. [06:55] They cost $499. For music and travel I'd buy them; if you take a lot of calls, look elsewhere.
This video is Maya’s review of the Aurora X2 wireless headphones after three weeks of daily use, for people considering them for music, travel, or calls. - 00:00 Maya introduces her review after three weeks of daily use. - 00:48 They weigh 254 grams, have an aluminium headband, and feature replaceable ear cushions, which she says is rare at this price. - 01:42 Maya measured 31 hours of battery life with noise cancelling on, versus 35 hours stated on the box. Ten minutes of charging provides about five hours. - 03:05 Noise cancelling works very well on planes and trains but struggles with voices in an open office. - 04:10 Call quality is the weak point: people said Maya sounded distant, even in a quiet room. - 05:30 The app allows EQ adjustment and pairing with two devices at once. - 06:55 They cost $499. Maya would buy them for music and travel, but recommends looking elsewhere if you take a lot of calls. Worth watching for: Maya’s three-week-use battery measurement and direct assessment of noise cancelling and call quality.
This is Maya's hands-on review of the Aurora X2 wireless headphones after three weeks of daily use, for anyone deciding whether to buy them. - [00:00] Maya introduces her review based on three weeks of daily use. - [00:48] Build: 254 grams, aluminium headband, and replaceable ear cushions, which is rare at this price. - [01:42] Battery: measured 31 hours with noise cancelling on, versus 35 hours claimed on the box; ten minutes of charging gives about five hours. - [03:05] Noise cancelling is very good on planes and trains but struggles with voices in an open office. - [04:10] Call quality is the weak point — people said she sounded distant, even in a quiet room. - [05:30] The app allows EQ adjustment and pairing with two devices at once. - [06:55] Price is $499. Verdict: good for music and travel, but if you take a lot of calls, look elsewhere. Worth watching for: Maya's real-world measurements (battery and noise cancelling) and her firsthand call-quality experience, which give context beyond the spec sheet.
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; how it's worded isn't judged.
The inputs: a one-bowl banana bread recipe, budgeting a first paycheck, a lecture on studying that sticks, a headphone review, as a timestamped outline, and the home office deduction's conditions. 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 YouTube summarizer
On YouTube, open the video, press “…more” under the title, then “Show transcript”. Select the transcript and copy it.
Paste it here. A free run takes up to 1,600 characters (about two minutes of talk); signed in, paste the whole thing.
Pick key points, a short paragraph or a timestamped outline, and the language.
Press “Summarize”. Without an account you see the start of the summary (1 free run a day, no account needed).
Tips for a better answer
Keep the timestamps
Paste the transcript with its timestamps (YouTube shows them by default) and pick the timestamped outline: every point gets a time you can jump to.
Long videos, one part at a time
For a long lecture, paste one section at a time and summarize each, or sign in and paste the whole transcript into one run.
Say what you're looking for
Add a line at the top such as “I only care about the part on pricing”, and the summary follows it.
Fix misheard names first
Captions get names and technical terms wrong. Fix the few that matter in the transcript before you paste it, and the summary will spell them right.
What a good summary looks like
Nothing the video didn't say
Every point in the summary should be something the video actually says. A summary that adds advice or facts the speaker never mentioned is guessing, however sensible it sounds.
Figures and steps kept exact
Numbers, names, prices and steps should come through exactly as spoken. If the video says “three sets of twelve”, the summary shouldn't say “a few sets”.
In the video's order
The points should follow the video's order, so you can find each one again. With timestamps pasted in, each point keeps its time.
Watch for caption mistakes
Auto-generated captions mishear names and jargon. If a word in the summary looks wrong, it's usually the transcript's spelling, not the model's.
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. 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, Claude Opus 5.5, 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 YouTube 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, Claude Opus 5.5, and GPT-6 Astra), with no card.
Can I just paste the YouTube link?
No. It reads the transcript you paste; it doesn't open links or watch videos. On YouTube, open the video, press “…more” under it, then “Show transcript”, and copy the text into the box.
How long a video can it summarize?
A free run takes up to 1,600 characters, about two minutes of talk. Signed in, paste the whole transcript: a very long one counts as more than one message once the conversation passes 64k tokens, and the message box says so before you send.
Can it summarize a video in another language?
Yes, if the transcript is in another language: it summarizes it in the language you pick, and keeps names and numbers as they are.
Can I get timestamps for each point?
Pick “Timestamped outline” and paste the transcript with its timestamps (YouTube shows them next to each line). Each point keeps the time it starts at, so you can jump to it.
Which AI is best for this, in your tests?
Every one of the 15 models we ran passed all 5 of our YouTube summarizer test inputs, so choose on speed and price: GLM 5.3 was quickest, at 1.3 s a run on average. GPT-6 Luna, the model a free run is on, took 6.6 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, and GLM, 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.