YouTube comments downloader

Get the comments.
Keep the context.

YouTube Video Transcript collects public comments and replies from videos, playlists, and channels. Take the conversation into a spreadsheet, your code, or your next research question.

Collect comments

No transcript needed. 30 free credits every 30 days with Google sign-in.

Top or Newest · Replies optional · JSON + CSV

Recorded thread excerpts

iPhone 14 Pro Review: This Will Be Copied!

Marques Brownlee

Top-level comment · 32K likes

just another one of the small marketing schemes that Apple does that works so well

A reply in its 498-reply thread

the main idea is the software-hardware integration, not the pill shaped camera

The top comment is not the whole conversation.

In the documented read of this review, the first ten replies showed more pushback than support for that top-level take. That finding is about the replies sampled, not all viewers.

Read the worked example

One video is a start.
A channel is a dataset.

Collect the conversation at the scale of your question. The same amount, order, and reply settings apply across the videos in your source.

One video

Look past the view count.

Collect the questions, reactions, and discussions beneath a review, tutorial, or interview. Start with a small sample or keep collecting within your budget.

A playlist

Compare the conversation.

Collect comments across a course, series, or curated set. Paste the playlist URL; the job finds its videos and applies your collection settings to each one.

A channel

Build an audience dataset.

Gather comments across a creator’s long-form uploads. Keep each video’s comments in separate files so you can compare topics without losing the source.

Up to 5,000 videos per job. Channels cover long-form uploads, not Shorts. Your credit limit applies to the entire collection.

Decide how deep to go.

OrderTop for ranked comments. Newest for recent ones.
Top follows YouTube’s ranking, not a strict sort by likes. Newest starts with recent comments. Choose the order that fits your question before the job begins.
Amount per video20, 100, 500, or all available.
Enter your own number of top-level comments per video, or choose All available. Collection stops when the amount is reached, YouTube has no more pages, or the job’s credit limit is used.
RepliesKeep the discussion attached.
Include replies to the top-level comments you collect. Replies are additional to your per-video amount, use the same credit budget, and retain their parent comment IDs in the export.
Example collection settings
Source
A playlist
Order
Newest first
Top-level comments per video
100
Include replies
Yes
Credit limit for the whole job
25

Aim for 100 top-level comments per video, plus their replies. Spend no more than 25 credits across the whole playlist. The budget may be reached before every video is covered.

Download comments as JSON and CSV.

Get a ZIP with JSON and CSV files for each video with saved comments. Read the CSV in a spreadsheet or use the JSON in your own analysis, with the source and reply relationships intact.

INSIDE YOUR COMMENTS ZIP

manifest.json

The source, collection settings, credits used, total saved comments, video statuses, and any stopping message.

For each video with collected comments

video-id.comments.csv

video-id.comments.json

Check the manifest before calling a collection complete. Videos without saved comments have a status entry, not an empty pair of data files.

What was said, and by whom.text · author
Read the comment text, identify recurring questions, and keep the author alongside each quote.
The engagement around it.likes · reply_count · published
Likes and publication times are YouTube’s display text, not exact numeric counts or timestamps.
Which reply belongs where.comment_id · is_reply · parent_comment_id
Join replies back to their parent comments when you group or analyze the conversation.
A way back to the source.url
Each row includes a YouTube comment link so you can revisit the original discussion.

You set the limit.
The job respects it.

One credit buys one nonempty page, usually about 20 top-level comments. Reply pages can be smaller. Empty pages and failed requests cost nothing.

Your chosen credit limit is reserved at the start. Collection stops when that limit is used, even with All available selected. Comments already saved remain yours to download.

If the job finishes or you cancel before the limit is used, the unused reservation is refunded automatically to the job’s original billing cycle.

Collect comments

30 free credits every 30 days. Paid plans start at $9/month for 1,000 credits, shared with transcripts. Compare plans.

EXAMPLE: FINISHING BELOW YOUR LIMIT
Credits reserved
25
Nonempty pages collected
9
Credits returned
16

You spend 9 credits, not 25. Billing is per page, not per comment or video. Page sizes vary, so a budget does not guarantee a fixed number of comments.

Website, API, and MCP.

The website handles background collections. API and MCP clients can retrieve comments page by page using the same credit balance.

A job in the website
Close the tab once it appears in History. Return to check the counts, cancel, or download the ZIP when collection stops. No manual paging needed.
Your code or AI chat
Use the comments API or MCP tools for individual pages and reply threads. Nonempty pages cost one credit each; no transcript purchase is required.

Before you collect.

Do I need to download a transcript first?

No. YouTube Video Transcript’s comments collector works independently. You only spend credits on nonempty comments pages, not on a transcript you do not need. A video can have available comments even when its captions are missing.

Does All available guarantee every comment?

No. It removes your per-video amount limit, but collection still ends when YouTube stops returning pages or the job reaches your credit limit. Deleted, hidden, and unavailable comments cannot be collected. Replies are included only if you enable them. The ZIP’s manifest records the collection settings, video statuses, and any stopping message so you can check coverage.

Will every video get the amount I requested?

The amount is a target for top-level comments on each video, not a guaranteed result. Some videos have fewer available comments. Videos are processed in sequence, and replies share the budget, so a limited budget can be spent before later videos are reached. Choose a smaller playlist or turn replies off when covering more videos matters most.

Can I collect comments from a channel I do not own?

Yes, when its videos and comments are publicly accessible. You do not need to connect the creator’s YouTube account. Collections support a video, playlist, or channel with up to 5,000 videos; channel collections exclude Shorts. If a source exceeds the limit, choose a smaller playlist. Disabled or inaccessible comments are reported, not bypassed.

Can I close the browser or cancel the collection?

Yes. Once the job appears in History, it runs on the server and does not need your tab open. Cancel it from History when you want to stop. Saved comments remain available in the ZIP, and unused reserved credits are refunded to the job’s original billing cycle. Download your results before the job’s 100-day retention period ends.

Can I try this on the free plan?

Yes. Sign in with Google for 30 free credits every 30 days, shared with transcripts, API, and MCP. Set a credit limit within your available balance. One nonempty page costs one credit; top-level pages usually contain about 20 comments, while reply page sizes vary. Empty pages and failed requests cost nothing. One comments collection can run at a time per account.

Does the collector analyze sentiment for me?

No. It collects and exports comments, authors, engagement text, and reply relationships. Use the CSV in a spreadsheet, process the JSON in your own code, or provide the data to an AI chat for analysis. Top comments are a ranked sample, not a poll of everyone who watched the video.

We use Google Analytics cookies and note which site referred you, so we know how people find us. Nothing personal, nothing sold. See our Privacy Policy.