2025 releases

Before updgrading to v6.0.0, update all relevant code to account for the following breaking changes. See the v6.0.0 release notes for full details.

• Incorporated the pinecone-plugin-records and pinecone-plugin-inference plugins into the pinecone package. If you are using these plugins, you must unstall them to use v6.0.0.
• Dropped support for Python 3.8, which has now reached official end of life, and added support for Python 3.13.
• Removed the explicit dependency on tqdm, which is used to provide a progress bar when upserting data into Pinecone. If tqdm is available in the environment, the Pinecone SDK will detect and use it, but tdqm is no longer required to run the SDK. Popular notebook platforms such as Jupyter and Google Colab already include tqdm in the environment by default, but if you are running small scripts in other environments and want to continue seeing progress bars, you will need to separately install the tqdm package.
• Removed some previously deprecated and rarely used keyword arguments (config, openapi_config, and index_api) to instead prefer dedicated keyword arguments for individual settings such as api_key, proxy_url, etc.

Before updgrading to v4.0.0, update all relevant code to account for the following breaking changes. See the v4.0.0 release notes for full details.

• embed method:

  • parameters now accepts Map<String, Object> instead of EmbedRequestParameters.
  • The Embeddings response class now has dense and sparse embeddings. You now must use getDenseEmbedding() or getSparseEmbedding(). For example, instead of embeddings.getData().get(0).getValues(), you would use embeddings.getData().get(0).getDenseEmbedding().getValues().

• rerank method:

  • documents now accepts List<Map<String, Object>> instead of List<Map<String, String>>.
  • parameters now accepts Map<String, Object> instead of Map<String, String>.

Before updgrading to v3.0.0, update all relevant code to account for the following breaking changes. See the v3.0.0 release notes for full details.

• embed operation:
  • EmbedParameters is no longer typed as a pointer.
• create_index operation:
  • CreateServerlessIndexRequest and CreatePodIndexRequest structs have been updated, and fields are now classified as pointers to better denote optionality around creating specific types of indexes: Metric, Dimension, VectorType, and DeletionProtection.
• Various data operation:
  • Values in the Vector type are now a pointer to allow flexibility when working with sparse-only indexes.

Before updgrading to v3.0.0, update all relevant code to account for the following breaking changes. See the v3.0.0 release notes for full details.

• embed operation:

  • The Embedding type has changed from a simple object to a discriminated union, supporting both DenseEmbedding and SparseEmbedding. New helper methods available on the Embedding type: IsDense and IsSparse for type checking, AsDense() and AsSparse() for type conversion, and Match() and Visit() for pattern matching.
  • The Parameters property now uses Dictionary<string, object?>? instead of EmbedRequestParameters.

• rerank operation:

  • The Document property now uses Dictionary<string, object?>? instead of Dictionary<string, string>?.
  • The Parameters property now uses Dictionary<string, object?>? instead of Dictionary<string, string>?.