mistralai-search-toolkit-plugins-qdrant

Qdrant backend for mistralai-search-toolkit.

Implements VectorStoreIndex, plus optional NavigableIndex and PatchableIndex.

Install

uv add mistralai-search-toolkit-plugins-qdrant
docker run -p 6333:6333 qdrant/qdrant

Usage

from mistralai.search.toolkit.embedding import CustomEmbeddingModel
from mistralai.search.toolkit.plugins.qdrant import (
    QdrantApp,
    QdrantCollectionSchema,
    QdrantConnectionConfig,
)
from mistralai.search.toolkit.search import VectorSearchQuery

schema = QdrantCollectionSchema(
    collection_name="docs",
    document_type=MyDoc,
    embedding_model=CustomEmbeddingModel(name="my-embedder", dimensions=1024),
)
app = QdrantApp([schema])
config = QdrantConnectionConfig(url="http://localhost:6333")

await app.create_collection(config, "docs")
store = app.get_search_index(config, "docs")
try:
    await store.index_document(doc)
    results = await store.search(VectorSearchQuery(embedding=vec, top_k=10))
finally:
    await store.aclose()

Pass an AsyncQdrantClient instead of a config to own the client. For tests:

QdrantConnectionConfig(path=":memory:")

One Qdrant point per chunk.

• Dense search. Reindex upserts, then drops stale points. Honors exclude_ids and max_candidates (hnsw_ef).
• Hybrid search. A query with both embedding and query fuses dense ranking with full-text over content via weighted RRF. Tune with QdrantSearchQuery. Fusion score is comparable only within one result set. distance comes from the dense side; lexical-only hits have distance=None.
• Navigation uses [start, end) offsets. grep is token-based: PHRASE ordered, TERM any order.
• patch_chunk / patch_document merge metadata key by key. None removes a key.

Custom chunk fields keep their name. Document fields get a document_ prefix.

from typing import Annotated
from mistralai.search.toolkit.plugins.qdrant import QdrantField

section: Annotated[str | None, QdrantField(name="section_title")] = None
internal: Annotated[str | None, QdrantField(ignore=True)] = None

Qdrant point ids are integers or UUIDs. Default UUID5 chunk ids pass through. Other ids map through uuid5; the original stays in payload id.

uv sync --group dev
uv run pytest tests/ -q