# Rumble (/docs/rumble)

Rumble videos, channels, comments, transcripts, and keyword search through one unified API

Rumble videos and shorts, channel uploads, full comment sections, spoken-word transcripts, and keyword search, all in the canonical `Post` and `Comment` shapes you get from YouTube and TikTok. Four of the five endpoints cost 1 credit; the transcript is premium at 10.

Base URL: `/v1/rumble/...`

  `video/comments` returns the whole top-level comment section in one response.
  There is no cursor and nothing to page, and replies are not expanded, so a
  thread's nesting is not available. What you get is every root comment.

## Quickstart

### 1. Find videos by keyword

```bash title="cURL"
curl "https://www.socialcrawl.dev/v1/rumble/search?query=space%20launch" \
  -H "x-api-key: YOUR_API_KEY"
```

### 2. Or walk one channel's uploads

```bash title="cURL"
curl "https://www.socialcrawl.dev/v1/rumble/channel/videos?handle=Bongino" \
  -H "x-api-key: YOUR_API_KEY"
```

### 3. Read computed fields

When an endpoint supports a computed field and the required source inputs are present, the unified response includes that optional field. Depending on the endpoint, optional fields can include `engagement_rate`, `language`, `content_category`, and `estimated_reach`. See [Computed fields](/docs/computed-fields.md) for formulas, clamping rules, and null semantics.

## Discovery

| Endpoint                        | Credits | What it returns                                                               | Key parameters              |
| ------------------------------- | ------- | ----------------------------------------------------------------------------- | --------------------------- |
| `GET /v1/rumble/search`         | 1       | Videos and shorts across all of Rumble: title, URL, thumbnail, channel, views | `query`, `cursor`           |
| `GET /v1/rumble/channel/videos` | 1       | One channel's own details plus its videos and shorts, cursor-paginated        | `handle` or `url`, `cursor` |

`channel/videos` needs one of the two identifiers. Sending neither is a `400` before billing.

## Video

| Endpoint                          | Credits | What it returns                                                                             | Key parameters |
| --------------------------------- | ------- | ------------------------------------------------------------------------------------------- | -------------- |
| `GET /v1/rumble/video`            | 1       | Title, description, thumbnail, channel, publish date, views, likes, dislikes, captions flag | `url`          |
| `GET /v1/rumble/video/comments`   | 1       | Every top-level comment, with author, timestamp, like and dislike counts, reply count       | `url`          |
| `GET /v1/rumble/video/transcript` | 10      | The spoken text of a video                                                                  | `url`          |

Read one video in full from its URL, then take the whole comment section in a second call.

```bash title="cURL"
curl "https://www.socialcrawl.dev/v1/rumble/video?url=https://rumble.com/v79xhhm-discovery-example.html" \
  -H "x-api-key: YOUR_API_KEY"

curl "https://www.socialcrawl.dev/v1/rumble/video/comments?url=https://rumble.com/v792vns-example.html" \
  -H "x-api-key: YOUR_API_KEY"
```

`video/transcript` reads Rumble's published captions. When a video has none, the transcript comes back empty and **the call is not charged**, so it is safe to attempt speculatively across a channel rather than pre-filtering. `video` reports whether captions exist if you would rather know before you ask.

## All endpoints

## Platform notes

**Comments arrive complete, but flat.** One response holds the entire top-level section, and replies are not expanded, so nesting is not available on this platform.

**Transcripts depend on published captions,** and cost nothing when there are none.

**Pagination is not uniform.** `search` and `channel/videos` page with the universal `cursor`; the three single-video endpoints do not paginate at all.

**Identifiers split three ways.** `channel/videos` takes `handle` **or** `url`; `video`, `video/comments`, and `video/transcript` all take the video `url`; `search` takes a `query`.

**Rumble feeds other surfaces too.** It is one of the sources behind [`/v1/search/everywhere`](/docs/search/everywhere.md) and a supported platform for `/v1/prism/video-intel`, which returns detail, stats, comments, and transcript for one video URL in a single call.

## Next steps

- [Pagination](/docs/pagination.md): Cursor and page walks, and what has_more actually means.
- [Credits](/docs/credits.md): What each tier costs and when a call is refunded.
- [API reference](/docs/api-reference.md): Every parameter and response field, endpoint by endpoint.
- [Twitch](/docs/twitch.md): Channels, videos, schedules, and clips.
- [Universal search](/docs/search/everywhere.md): One keyword across Rumble and every other source at once.