Collections API
Create, configure, and manage vector collections with flexible schema definitions.
Create Collection
POST
/api/v1/collections
Schema Configuration
Define the structure and properties of your vector collection:
{
"name": "my_collection",
"schema": {
"fields": [
{
"name": "id",
"type": "int64",
"is_primary": true
},
{
"name": "vector",
"type": "float_vector",
"dimension": 512
},
{
"name": "metadata",
"type": "json"
}
]
},
"index_params": {
"metric_type": "L2",
"index_type": "HNSW",
"params": {
"M": 16,
"efConstruction": 200
}
}
}
Performance Optimization
- Choose the right index: HNSW for high accuracy, IVF for memory efficiency
- Optimize dimensions: Higher dimensions provide more precision but slower queries
- Batch operations: Insert vectors in batches for better throughput