lfnovo/open-notebook · error · HTTPException

No embedding model configured. Please configure one in the M

Error message

No embedding model configured. Please configure one in the Models section.

What it means

400 from POST /api/embed when no default embedding model is configured in the Models section. Embedding requires a default embedding model; model_manager.get_embedding_model() returning falsy means none is set, so the request is rejected before any work.

Source

Thrown at api/routers/embedding.py:22

from api.command_service import CommandService
from api.models import EmbedRequest, EmbedResponse
from open_notebook.ai.models import model_manager
from open_notebook.domain.notebook import Note, Source
from open_notebook.exceptions import (
    NotFoundError,
    OpenNotebookError,
)

router = APIRouter()


@router.post("/embed", response_model=EmbedResponse)
async def embed_content(embed_request: EmbedRequest):
    """Embed content for vector search."""
    try:
        # Check if embedding model is available
        if not await model_manager.get_embedding_model():
            raise HTTPException(
                status_code=400,
                detail="No embedding model configured. Please configure one in the Models section.",
            )

        item_id = embed_request.item_id
        item_type = embed_request.item_type.lower()

        # Validate item type
        if item_type not in ["source", "note"]:
            raise HTTPException(
                status_code=400, detail="Item type must be either 'source' or 'note'"
            )

        # Branch based on processing mode
        if embed_request.async_processing:
            # ASYNC PATH: Submit command for background processing
            logger.info(f"Using async processing for {item_type} {item_id}")

View on GitHub (pinned to a7de90d38a)

Solutions

Open the Models section in the UI (or GET /api/models) and set a default embedding model
If no embedding-capable model exists, add one (built-in or via a configured credential) first
Ensure the embedding model's backing credential is valid (has a working API key)
Retry the embed request after the default is set

Defensive patterns

Strategy: validation

Validate before calling

const models = await api.listModels();
const hasEmbeddingDefault = models.some(m => m.isDefault && m.type === 'embedding');
if (!hasEmbeddingDefault) throw new Error('Configure a default embedding model first');

Try / catch

try {
  await api.embed({ item_id: id, item_type: 'source', async_processing: true });
} catch (e) {
  if (e.status === 400 && /embedding model/i.test(e.detail)) promptModelSetup();
  throw e;
}

Prevention

Configure a default embedding model as part of first-run setup
Re-check model defaults after deleting or changing embedding models

When it happens

Trigger: Calling POST /api/embed (e.g. saving a note or triggering vector search) on a fresh install where no embedding model was ever selected, or after the configured embedding model was deleted.

Common situations: Fresh Open Notebook setup where the user skipped the Models configuration step, or removed/changed the default embedding model and stale clients still submit embed requests.

Related errors

AI-assisted analysis of lfnovo/open-notebook@a7de90d38a (2026-08-27). Data as JSON: /api/errors/08202eaf1d5586a1. Report an issue: GitHub.