Search documentation
Search pages, API reference sections, and guide headings.
Summary
Indexes are top-level collections. Media resources include uploaded or registered videos, audio files, images, segment listings, batch helpers, and playground-only media.
Index endpoints
Indexes are the collection boundary for video-to-vector embeddings and multimodal media search. Prompt runs can add
transcription, image embeddings, and metadata_text so later search requests can use both semantic vectors and exact
structured metadata.
| Method | Path | Purpose |
|---|---|---|
POST | /api/v2/indexes | Create an index |
GET | /api/v2/indexes | List indexes, including defaults when requested |
GET | /api/v2/indexes/{index_id} | Retrieve one index |
DELETE | /api/v2/indexes/{index_id} | Delete an index |
GET | /api/v2/indexes/{index_id}/videos | Paginate media in an index |
GET | /api/v2/indexes/{index_id}/prompt-runs | Paginate runs for an index |
Media endpoints
| Method | Path | Purpose |
|---|---|---|
POST | /api/v2/videos | Register media from an existing video_uri |
POST | /api/v2/videos/upload | Upload a video, audio file, or image |
GET | /api/v2/videos/{video_id} | Retrieve a media item |
DELETE | /api/v2/videos/{video_id} | Delete a media item |
GET | /api/v2/videos/{video_id}/segments | Paginate segments |
GET | /api/v2/videos/{video_id}/prompt-runs | List runs touching a media item |
POST | /api/v2/videos/batch | Batch fetch media with detail fields |
POST | /api/v2/videos/batch/status | Batch fetch processing status snapshots |
POST | /api/v2/videos/batch/segments | Batch fetch segment lists |
Playground media
Playground media is index-less media owned by the current user.
| Method | Path | Purpose |
|---|---|---|
GET | /api/v2/playground/videos | Paginate playground media |
Example: create an index
curl -X POST /api/v2/indexes \
-H "X-API-Key: sk_live_..." \
-H "Content-Type: application/json" \
-H "Idempotency-Key: index-create-archive-2026-04-20" \
-d '{"name":"Broadcast archive"}'
Example: upload media into an index
curl -X POST /api/v2/videos/upload \
-H "X-API-Key: sk_live_..." \
-F "file=@episode-001.mp4" \
-F "title=Episode 001" \
-F "index_id=idx_archive"
Segment retrieval
GET /api/v2/videos/{video_id}/segments supports:
run_idto pin results to one prompt runlatest_run=trueto retrieve the most recent run-backed segments- cursor pagination
Use this endpoint when you need the canonical segment records rather than ranked search results.
SDK equivalents
client.indexes.create,client.indexes.list,client.indexes.list_videosclient.videos.upload,client.videos.retrieve,client.videos.list_segmentsclient.videos.batch_retrieve,client.videos.batch_status,client.videos.batch_segmentsclient.videos.list_playground
Use search endpoints to find evidence. Use media and segment endpoints when you already know the media boundary and need canonical records, status snapshots, or pagination.
Related documentation
VideoVector organizes media workflows around indexes, prompts, prompt runs, search, and delivery resources. This page explains how those public entities fit together.
Prompt runs are the execution boundary for extraction. The API exposes run creation, status inspection, cancellation, segment results, media-wide synthesis, failed-segment manifests, and debug-oriented LLM call access.
These examples show how the SDK maps common implementation flows directly onto the platform model, from prompt design and execution to search, connectors, imports, exports, and webhooks.