ValueError
qdrant_client.http.exceptions.ValueError
Stack trace
ValueError: Vector dimension mismatch: expected 1536, got 512
File "/usr/local/lib/python3.9/site-packages/qdrant_client/http/api/points_api.py", line 123, in upsert
raise ValueError(f"Vector dimension mismatch: expected {expected_dim}, got {actual_dim}") Why it happens
Qdrant collections are created with a fixed vector dimension. When inserting points, the vector length must exactly match this dimension. If the inserted vector's dimension differs, Qdrant raises a ValueError to prevent inconsistent data storage.
Detection
Check vector length before insertion by asserting len(vector) == collection_vector_size or catch ValueError on insert and log the vector dimensions for debugging.
Causes & fixes
The vector being inserted has a different dimension than the Qdrant collection's configured dimension.
Ensure the vector embedding generation matches the collection's dimension or recreate the collection with the correct dimension.
Using multiple embedding models with different output sizes but inserting into the same Qdrant collection.
Use separate collections for each embedding dimension or unify embedding models to produce vectors of the same size.
The collection was created with a default or incorrect dimension setting not matching the actual vectors.
Delete and recreate the collection with the correct vector size parameter before inserting data.
Code: broken vs fixed
from qdrant_client import QdrantClient
client = QdrantClient()
collection_name = "my_collection"
vector = [0.1] * 512 # Vector dimension 512
# This line raises ValueError due to dimension mismatch
client.upsert(collection_name=collection_name, points=[{'id': 1, 'vector': vector}])import os
from qdrant_client import QdrantClient
client = QdrantClient(api_key=os.environ.get('QDRANT_API_KEY'))
collection_name = "my_collection"
vector = [0.1] * 1536 # Fixed vector dimension to match collection
# Fixed: vector dimension matches collection dimension
client.upsert(collection_name=collection_name, points=[{'id': 1, 'vector': vector}])
print("Insert succeeded with matching vector dimension.")Workaround
Catch the ValueError on insert, log the vector dimension mismatch, and programmatically resize or regenerate vectors to the correct dimension before retrying insertion.
Prevention
Always create Qdrant collections with a known fixed vector dimension and ensure all embedding vectors conform to this dimension before insertion, ideally by standardizing embedding model usage.