--- title: embeddings | CodeWeaver Docs description: API reference for codeweaver.core.types.embeddings url: "https://docs.knitli.com/api/core/types/embeddings" type: static generatedAt: "2026-04-17T17:21:08.168Z" --- # embeddings [Open in ChatGPT](https://chatgpt.com/?q=Read%20https%3A%2F%2Fdocs.knitli.com%2Fcodeweaver%2Fapi%2Fcore%2Ftypes%2Fembeddings%2F.%20I%20want%20to%20ask%20questions%20about%20it.)[Open in Claude](https://claude.ai/new?q=Read%20https%3A%2F%2Fdocs.knitli.com%2Fcodeweaver%2Fapi%2Fcore%2Ftypes%2Fembeddings%2F.%20I%20want%20to%20ask%20questions%20about%20it.)[View in Markdown](/codeweaver/api/core/types/embeddings.md) [Share on LinkedIn](https://www.linkedin.com/sharing/share-offsite/?url=https%3A%2F%2Fdocs.knitli.com%2Fcodeweaver%2Fapi%2Fcore%2Ftypes%2Fembeddings%2F)[Share on X](https://x.com/intent/tweet?url=https%3A%2F%2Fdocs.knitli.com%2Fcodeweaver%2Fapi%2Fcore%2Ftypes%2Fembeddings%2F&text=embeddings)[Share on Threads](https://threads.net/intent/post?url=https%3A%2F%2Fdocs.knitli.com%2Fcodeweaver%2Fapi%2Fcore%2Ftypes%2Fembeddings%2F&text=embeddings)[Share on Bluesky](https://bsky.app/intent/compose?text=embeddings%20https%3A%2F%2Fdocs.knitli.com%2Fcodeweaver%2Fapi%2Fcore%2Ftypes%2Fembeddings%2F)[Share on Facebook](https://www.facebook.com/sharer/sharer.php?u=https%3A%2F%2Fdocs.knitli.com%2Fcodeweaver%2Fapi%2Fcore%2Ftypes%2Fembeddings%2F)[Share on Reddit](https://reddit.com/submit?url=https%3A%2F%2Fdocs.knitli.com%2Fcodeweaver%2Fapi%2Fcore%2Ftypes%2Fembeddings%2F&title=embeddings)[Share on Hacker News](https://news.ycombinator.com/submitlink?u=https%3A%2F%2Fdocs.knitli.com%2Fcodeweaver%2Fapi%2Fcore%2Ftypes%2Fembeddings%2F&t=embeddings)[Share on Email](mailto:?subject=embeddings&body=https%3A%2F%2Fdocs.knitli.com%2Fcodeweaver%2Fapi%2Fcore%2Ftypes%2Fembeddings%2F)[Share on WhatsApp](https://wa.me/?text=embeddings%20https%3A%2F%2Fdocs.knitli.com%2Fcodeweaver%2Fapi%2Fcore%2Ftypes%2Fembeddings%2F)[Share on Telegram](https://t.me/share/url?url=https%3A%2F%2Fdocs.knitli.com%2Fcodeweaver%2Fapi%2Fcore%2Ftypes%2Fembeddings%2F&text=embeddings) # `codeweaver.core.types.embeddings` [Section titled “codeweaver.core.types.embeddings”](#codeweavercoretypesembeddings) Core embedding type definitions for CodeWeaver. ## Class: `ChunkEmbeddings` [Section titled “Class: ChunkEmbeddings”](#class-chunkembeddings) Model representing the embeddings associated with a specific CodeChunk. Uses a dynamic dictionary to support arbitrary named embeddings (intents). Common keys include: “primary”, “sparse”, “backup”, “summary”, “ast”. ### Method: `add` [Section titled “Method: add”](#method-add) ** ``` add() ``` Add an EmbeddingBatchInfo to the ChunkEmbeddings. Args: embedding_info: The embedding information to add (intent is extracted from this object). Returns: ChunkEmbeddings: A new ChunkEmbeddings instance with the added embedding. Raises: ValueError: If embedding already exists for this intent or chunk IDs don’t match. ### Method: `get_model_by_vector_intent` [Section titled “Method: get_model_by_vector_intent”](#method-get_model_by_vector_intent) ** ``` get_model_by_vector_intent() ``` Get the model name used for a specific embedding intent, if it exists. ### Method: `update` [Section titled “Method: update”](#method-update) ** ``` update() ``` Update or replace an EmbeddingBatchInfo in the ChunkEmbeddings. Unlike `add()`, this method will replace existing embeddings of the same intent. This is useful for re-embedding scenarios where we want to update embeddings. Args: embedding_info: The embedding information to update/replace (intent is extracted from this object). Returns: ChunkEmbeddings: A new ChunkEmbeddings instance with the updated embedding. Raises: ValueError: If chunk IDs don’t match. ## Class: `CodeWeaverSparseEmbedding` [Section titled “Class: CodeWeaverSparseEmbedding”](#class-codeweaversparseembedding) NamedTuple representing sparse embedding with indices and values. Sparse embeddings are represented as two parallel arrays: - indices: positions in the vocabulary that have non-zero values - values: weights/importance scores for those positions This format is used by SPLADE, SparseEncoder and similar models. ### Method: `to_tuple` [Section titled “Method: to_tuple”](#method-to_tuple) ** ``` to_tuple() ``` Convert to a CodeWeaverSparseEmbedding with tuples for indices and values. ## Class: `DataType` [Section titled “Class: DataType”](#class-datatype) Enum representing the data type of embedding vectors. ### Method: `cross_walk` [Section titled “Method: cross_walk”](#method-cross_walk) ** ``` cross_walk() ``` Get the corresponding data type string for crosswalks. ## Class: `EmbeddingBatchInfo` [Section titled “Class: EmbeddingBatchInfo”](#class-embeddingbatchinfo) BasedModel representing metadata about a CodeChunk’s embedding within a batch. ### Method: `create_dense` [Section titled “Method: create_dense”](#method-create_dense) ** ``` create_dense() ``` Create EmbeddingBatchInfo for dense embeddings. ### Method: `create_sparse` [Section titled “Method: create_sparse”](#method-create_sparse) ** ``` create_sparse() ``` Create EmbeddingBatchInfo for sparse embeddings. Args: batch_id: Unique identifier for the batch batch_index: Index within the batch chunk_id: Unique identifier for the chunk model: Model name embeddings: CodeWeaverSparseEmbedding with indices and values intent: Intent/purpose of this embedding (default: “sparse”) dimension: Dimensionality of the embedding (always 0 for sparse) dtype: Data type of the embedding values ## Class: `EmbeddingKind` [Section titled “Class: EmbeddingKind”](#class-embeddingkind) Enum representing the kind of embedding. ## Class: `QueryResult` [Section titled “Class: QueryResult”](#class-queryresult) Multi-vector embedding result from an embedding query. Stores embeddings from multiple intents (primary, backup, sparse, etc.) in a dictionary keyed by intent name. ### Method: `get` [Section titled “Method: get”](#method-get) ** ``` get() ``` Get embedding for an intent with optional default.