Search branch files with a vector
Search branch files with a vector
Search branch content using a pre-computed embedding vector. Generate embeddings client-side using OpenAI, Anthropic, or another embedding provider.
POST
Search branch files with a vector
Authorizations
Use a JWT bearer token or a Veryfront API key in the Authorization header.
Path Parameters
Project ID, slug, or domain from the authenticated context.
Branch name or ID. Use main to search the default project state.
Body
application/json
Pre-computed query embedding vector.
Required array length:
768 - 4096 elementsEmbedding vector dimension: 768, 1024, 1536, 3072, or 4096.
Available options:
768 Maximum search results to return, from 1 to 100.
Required range:
1 <= x <= 100Opaque cursor returned from page_info.next.
Minimum similarity score for returned semantic matches.
Required range:
0 <= x <= 1Response
Search results.