Migrate from Azure AI Search to Qdrant

  • Service endpoint — e.g. https://your-service.search.windows.net
  • Index name — the index containing your vectors
  • Admin API key — used to authenticate with the Azure AI Search REST API

Concept Mapping

Azure AI SearchQdrantNotes
IndexCollectionOne-to-one mapping
DocumentPointEach document becomes a point
Vector field (Collection(Edm.Single) with dimensions)Named vectorEach vector field in the index becomes a named vector on the same point
Retrievable document fieldsPayloadNon-vector, retrievable fields (excluding @search.* metadata) become payload
cosineCosineDirect mapping; also the fallback for any unrecognized metric
dotProductDotDirect mapping
euclideanEuclidDirect mapping

Run the Migration

docker run --net=host --rm -it registry.cloud.qdrant.io/library/qdrant-migration azure \
    --azure.endpoint 'https://your-service.search.windows.net' \
    --azure.index 'your-index' \
    --azure.api-key 'your-azure-admin-api-key' \
    --qdrant.url 'https://your-instance.cloud.qdrant.io:6334' \
    --qdrant.api-key 'your-qdrant-api-key' \
    --qdrant.collection 'your-collection'

All Azure AI Search-Specific Flags

FlagRequiredDescription
--azure.endpointYesAzure AI Search service endpoint
--azure.indexYesIndex to migrate
--azure.api-keyYesAdmin API key for authentication
--azure.api-versionNoREST API version (default 2026-04-01)

Qdrant-Side Options

FlagDefaultDescription
--qdrant.id-field__id__Payload field name for original Azure AI Search document keys

Next Steps

After migration, verify your data arrived correctly with the Migration Verification Guide.

Was this page useful?

Thank you for your feedback! 🙏

We are sorry to hear that. 😔 You can edit this page on GitHub, or create a GitHub issue.