# Tumblr Search API (tumblr.search)

- Canonical page: [https://everyinfra.com/en/tumblr-api/search](https://everyinfra.com/en/tumblr-api/search?utm_medium=ai_doc&utm_campaign=page_md)
- Endpoint: POST https://api.everyinfra.com/api/v1/social with {"platform": "tumblr", "action": "search", "params": {...}}
- Price: $0.036 per call ($36 per 1,000 requests); failed calls and empty results are not charged
- Returns: list, default 20, max 200 per call
- Mode: sync; add "mode": "async" for slow tasks

Tumblr Search finds public Tumblr posts by `keyword` and returns each post's text, link, post type, tags, note count and publish time, with the blog's username, name and link. Fan community researchers, brand teams and content strategists use it to see how a topic, fandom, product or visual style is discussed on Tumblr, to collect posts for content or sentiment research, or to find the blogs that post most about a subject.

`keyword` is a search term, not a tag. Narrow results to one kind of post with `post_type`, choosing any, text, photo, video, quote, link or audio, one value per request. Blogs mix these types and Tumblr lays each out differently, so the type is worth keeping on every row you store. `tags` comes back as an empty list rather than null when a post has none, and `posted_at` is in UTC.

## Health

- Live health (24h percent, hourly and daily trends): https://api.everyinfra.com/api/v1/social/availability?platform=tumblr&action=search

## Parameters

| name | type | required | default | description |
|---|---|---|---|---|
| keyword | not declared | yes | not declared | Public post search term for `search`. The compatibility keyword input for `tag` actually queries by tag and cannot be treated as an identical search entry point. |
| post_type | not declared | no | not declared | Single post type for `search`, `tag`, or `user_posts`: `any` for unrestricted, `text`, `photo`, `video`, `quote`, `link`, or `audio`. This is not an array, and `profile` does not use this filter. |
| limit | integer (1–200) | no | 20 | The most rows to return in this call, an integer from 1 to 200. The default applies when you leave it out, and a call can return fewer rows. Larger values return 422 before any charge. |

## Response fields (items in `results`)

| field | meaning |
|---|---|
| id | The record's unique ID on the platform. Unique only within one platform; IDs can repeat across platforms. |
| url | Link to the record on the platform. |
| text | Body text with HTML removed, returned in full and never cut short (on a few platforms the list results themselves carry only an excerpt; the platform's field notes say so). |
| post_type | platform-native meaning |
| author_username | Author's username (@handle), usually usable in URLs. |
| author_name | Author's display name. |
| author_url | Link to the author's profile on the platform. |
| note_count | platform-native meaning |
| posted_at | Publish time in ISO 8601, UTC (for example 2026-08-07T12:34:56+00:00). Non-standard strings from the platform are passed through as is, so parse defensively. null means the platform does not disclose it. |
| platform | Source platform identifier, the same as platform in the request (for example xiaohongshu, tiktok). |
| tags | Array of tags. An empty array, not null, when there are none. |

## Example

```bash
curl -X POST 'https://api.everyinfra.com/api/v1/social' \
  -H "Authorization: Bearer $EVERYINFRA_API_KEY" \
  -H 'Content-Type: application/json' \
  --data-raw '{
  "platform": "tumblr",
  "action": "search",
  "params": {
    "keyword": "film photography"
  }
}'
```

```python
import os
import requests

response = requests.post(
    "https://api.everyinfra.com/api/v1/social",
    headers={"Authorization": f"Bearer {os.environ['EVERYINFRA_API_KEY']}"},
    json={
        "platform": "tumblr",
        "action": "search",
        "params": {
            "keyword": "film photography",
        },
    },
    timeout=120,
)
print(response.status_code)
print(response.json())
```

## Response example

Real response to the example request above, captured 2026-09-26T08:56:04Z. `results` and `count` are real with personal information (names, handles, avatars, profile links, contact details) removed; the list keeps the first 2 of 20. `id`, `billing` and `quota` differ on every call, so only their shape is shown.

```text
{
  "id": "req_…",
  "platform": "tumblr",
  "action": "search",
  "results": [
    {
      "id": "•••",
      "url": "https://www.tumblr.com/blog/view/heyscola/828814013050798080",
      "text": "Hudson Yards, 2026",
      "post_type": "text",
      "author_username": "•••",
      "author_name": "•••",
      "author_url": "https://www.tumblr.com/•••",
      "note_count": 0,
      "posted_at": "2026-09-26T10:30:37+00:00",
      "platform": "tumblr",
      "tags": [
        "black and white",
        "manhattan",
        "35mm"
      ]
    },
    {
      "id": "•••",
      "url": "https://capturedbykat.tumblr.com/post/828810369369882624/above-the-rio-grande-ii-orilla-verde",
      "text": "// Above the Rio Grande II // Orilla Verde Recreation Area, NM Olympus Infinity Stylus, Seattle Film Works expired film, August 2026",
      "post_type": "text",
      "author_username": "•••",
      "author_name": "•••",
      "author_url": "https://capturedbykat.tumblr.com/•••",
      "note_count": 0,
      "posted_at": "2026-09-26T09:32:42+00:00",
      "platform": "tumblr",
      "tags": [
        "rio grande",
        "orilla verde",
        "film photography"
      ]
    }
  ],
  "count": 20,
  "billing": {
    "charged": …,
    "charged_usd": …,
    "balance_after_usd": …,
    …
  },
  "quota": { … }
}
```

## FAQ

### Is `keyword` the same as searching a tag?

No. `keyword` is a search term matched against public posts, while a tag search reads the posts filed under one tag. For that, call the Tumblr Tag capability with the tag name, with or without the #.

### How do I read `note_count`?

It is Tumblr's single interaction total: likes and reblogs are merged into notes and cannot be separated. A high count means a post drew many likes, many reblogs or both, so compare it only with other note counts, not with like or share figures from other platforms.

### How do I read more from a blog I found?

Pass the row's `author_username`, the blog's name, to the Tumblr Posts by user capability, which lists that blog's public posts and accepts the same `post_type` filter. `author_url` opens the blog itself in a browser.

## Also

- Capability catalog (no key): https://api.everyinfra.com/api/v1/social/catalog?platform=tumblr
- OpenAPI: https://everyinfra.com/openapi.json · Agents guide: https://everyinfra.com/agents.md
