Overview
Pinecone is a managed vector database service designed for production-scale similarity search. It’s cloud-only and supports both dense and sparse vectors with automatic scaling. Provider Class:PineconeProviderConfig Class:
PineconeConfig
Dependencies
Examples
Parameters
| Parameter | Type | Description | Default | Source |
|---|---|---|---|---|
collection_name | str | Name of the collection | "default_collection" | Base |
vector_size | int | Dimension of vectors | Required | Base |
distance_metric | DistanceMetric | Similarity metric (COSINE, EUCLIDEAN, DOT_PRODUCT) | COSINE | Base |
recreate_if_exists | bool | Recreate collection if it exists | False | Base |
default_top_k | int | Default number of results | 10 | Base |
default_similarity_threshold | Optional[float] | Minimum similarity score | None | Base |
api_key | SecretStr | Pinecone API key | Required | Specific |
spec | Optional[Union[Dict, ServerlessSpec, PodSpec]] | Index specification | None | Specific |
environment | Optional[str] | Environment/region (backward compatibility) | None | Specific |
namespace | Optional[str] | Namespace for data isolation | None | Specific |
metric | Literal['cosine', 'euclidean', 'dotproduct'] | Distance metric | 'cosine' | Specific |
pods | Optional[int] | Number of pods (PodSpec) | None | Specific |
pod_type | Optional[str] | Pod type specification | None | Specific |
replicas | Optional[int] | Number of replicas (PodSpec) | None | Specific |
shards | Optional[int] | Number of shards (PodSpec) | None | Specific |
use_sparse_vectors | bool | Enable sparse vector support | False | Specific |
batch_size | int | Batch size for upsert operations | 100 | Specific |
timeout | Optional[int] | Request timeout in seconds | None | Specific |

