Create an index
Create a Pinecone index. This is where you specify the measure of similarity, the dimension of vectors to be stored in the index, which cloud provider you would like to deploy with, and more.
For guidance and examples, see Create an index.
Authorizations
Body
The configuration needed to create a Pinecone index.
The name of the index. Resource name must be 1-45 characters long, start and end with an alphanumeric character, and consist only of lower case alphanumeric characters or '-'.
1 - 45"example-index"
The spec object defines how the index should be deployed.
For serverless indexes, you set only the cloud and region where the index should be hosted. For pod-based indexes, you set the environment where the index should be hosted, the pod type and size to use, and other index characteristics. For BYOC indexes, you set the environment name provided to you during onboarding.
The dimensions of the vectors to be inserted in the index.
1 <= x <= 200001536
The distance metric to be used for similarity search. You can use 'euclidean', 'cosine', or 'dotproduct'. If the 'vector_type' is 'sparse', the metric must be 'dotproduct'. If the vector_type is dense, the metric defaults to 'cosine'.
cosine, euclidean, dotproduct Whether deletion protection is enabled/disabled for the index.
disabled, enabled Custom user tags added to an index. Keys must be 80 characters or less. Values must be 120 characters or less. Keys must be alphanumeric, '', or '-'. Values must be alphanumeric, ';', '@', '', '-', '.', '+', or ' '. To unset a key, set the value to be an empty string.
{ "tag0": "val0", "tag1": "val1" }
The index vector type. You can use 'dense' or 'sparse'. If 'dense', the vector dimension must be specified. If 'sparse', the vector dimension should not be specified.
Response
The IndexModel describes the configuration and status of a Pinecone index.
The name of the index. Resource name must be 1-45 characters long, start and end with an alphanumeric character, and consist only of lower case alphanumeric characters or '-'.
1 - 45"example-index"
The distance metric to be used for similarity search. You can use 'euclidean', 'cosine', or 'dotproduct'. If the 'vector_type' is 'sparse', the metric must be 'dotproduct'. If the vector_type is dense, the metric defaults to 'cosine'.
cosine, euclidean, dotproduct The URL address where the index is hosted.
"semantic-search-c01b5b5.svc.us-west1-gcp.pinecone.io"
{
"pod": {
"environment": "us-east-1-aws",
"metadata_config": {
"indexed": ["genre", "title", "imdb_rating"]
},
"pod_type": "p1.x1",
"pods": 1,
"replicas": 1,
"shards": 1
}
}
{
"ready": true,
"state": "ScalingUpPodSize"
}
The index vector type. You can use 'dense' or 'sparse'. If 'dense', the vector dimension must be specified. If 'sparse', the vector dimension should not be specified.
The dimensions of the vectors to be inserted in the index.
1 <= x <= 200001536
Whether deletion protection is enabled/disabled for the index.
disabled, enabled Custom user tags added to an index. Keys must be 80 characters or less. Values must be 120 characters or less. Keys must be alphanumeric, '', or '-'. Values must be alphanumeric, ';', '@', '', '-', '.', '+', or ' '. To unset a key, set the value to be an empty string.
{ "tag0": "val0", "tag1": "val1" }
The embedding model and document fields mapped to embedding inputs.
{
"field_map": { "text": "your-text-field" },
"metric": "cosine",
"model": "multilingual-e5-large",
"read_parameters": { "input_type": "query", "truncate": "NONE" },
"write_parameters": { "input_type": "passage" }
}