API Reference

Refresh Embeddings

Re-embed this end user's memories after an embedding-model change. Returns immediately and continues in the background.

Endpoint

POST/memory/refresh

202 Accepted

Takes no body. Scope comes from the API key, project header and end-user header.

Trigger a refresh

import os
from memorysync import MemorySyncClient

client = MemorySyncClient(
    api_key=os.environ["MEMORYSYNC_API_KEY"],
    base_url="https://api.memorysync.io",
    project_id=os.environ["MEMORYSYNC_PROJECT_ID"],
    end_user_id="usr_7f3a9c2e",
)

result = client.refresh()

Result handling

When to use it

After the embedding model for your search space changes.
When retrieval quality drops in a way that points at stale vectors rather than content.
Not as routine maintenance — normal writes embed themselves.

Errors and next action

For 429, wait before requesting another refresh; the work is idempotent but repeated requests queue repeated work.

Safety notes

Query Memory

Verify retrieval after a refresh.

Memory Status

Check per-memory processing state.

← Previous

Compose

Memory Graph