Migrate from FAISS to Qdrant What You Need from FAISS Index file path — path to your FAISS index fileDistance metric — the metric used when the index was built (l2, inner product, etc.)Important: Only non-quantized FAISS index types are supported. Quantized indexes (e.g., IndexIVFPQ) do not store the original vectors and cannot be migrated.Supported Index Types FAISS Index Type Supported Notes IndexFlatL2Yes Maps to euclid distance IndexFlatIPYes Maps to dot distance IndexHNSWFlatYes Full vectors are stored IndexIVFFlatYes Full vectors are stored IndexIVFPQNo Quantized — original vectors not stored IndexPQNo Quantized — original vectors not stored
Concept Mapping FAISS Qdrant Notes Index Collection One-to-one mapping Vector (by position) Point Position in index becomes point ID
Run the Migration docker run --net= host --rm -it \
-v /path/to/your/index:/data \
registry.cloud.qdrant.io/library/qdrant-migration faiss \
--faiss.index-path '/data/your-index.index' \
--qdrant.url 'https://your-instance.cloud.qdrant.io:6334' \
--qdrant.api-key 'your-qdrant-api-key' \
--qdrant.collection 'your-collection' \
--qdrant.distance-metric cosine
All FAISS-Specific Flags Flag Required Description --faiss.index-pathYes Path to the FAISS index file (inside the container)
Qdrant-Side Options Flag Default Description --qdrant.distance-metriccosineDistance metric: cosine, dot, euclid, or manhattan
Gotchas No metadata: FAISS stores only vectors, so every point has an empty payload. Point IDs are the vectors’ positions in the index, which you can use to join an external metadata store.Next Steps After migration, verify your data arrived correctly with the Migration Verification Guide .
Ready to try Qdrant?