@radzor/embeddings-store

In-memory vector store for embeddings with cosine similarity search. Supports OpenAI and Ollama embedding providers. Ideal for RAG pipelines and semantic search.

AI & MLv0.1.0typescriptpythonUniversalembeddingsvectorragsimilaritysearchopenaisemanticby Radzor

Install

View source on GitHub →

$npx radzor@latest add embeddings-store

⚠ Constraints: In-memory store — data does not persist across restarts. For production, export/import the store or use a dedicated vector DB.

Inputs

Name	Type	Default	Description
provider*	'openai' \| 'ollama'	—	Embedding provider.
apiKey	string	—	API key (required for OpenAI).OPENAI_API_KEY
model	string	—	Embedding model (e.g. text-embedding-3-small).
baseUrl	string	—	Override base URL for self-hosted.

provider*'openai' | 'ollama'

Embedding provider.

apiKeystring

{ "llm": { "constraints": "In-memory store — data does not persist across restarts. For production, export/import the store or use a dedicated vector DB.", "usageExamples": "llm/examples.md", "integrationPrompt": "llm/integration.md" }, "name": "@radzor/embeddings-store", "tags": [ "embeddings", "vector", "rag", "similarity", "search", "openai", "semantic" ], "events": [ { "name": "onStored", "payload": { "id": "string", "dimensions": "number" }, "description": "Fired when an embedding is stored." }, { "name": "onSearchComplete", "payload": { "query": "string", "resultCount": "number" }, "description": "Fired after a search." }, { "name": "onError", "payload": { "code": "string", "message": "string" }, "description": "Fired on errors." } ], "inputs": [ { "name": "provider", "type": "'openai' | 'ollama'", "required": true, "description": "Embedding provider." }, { "name": "apiKey", "type": "string", "envVar": "OPENAI_API_KEY", "required": false, "description": "API key (required for OpenAI)." }, { "name": "model", "type": "string", "required": false, "description": "Embedding model (e.g. text-embedding-3-small)." }, { "name": "baseUrl", "type": "string", "required": false, "description": "Override base URL for self-hosted." } ], "radzor": "1.0.0", "actions": [ { "name": "add", "params": [ { "name": "id", "type": "string", "description": "Unique document ID." }, { "name": "text", "type": "string", "description": "Text to embed and store." }, { "name": "metadata", "type": "Record<string, any>", "description": "Optional metadata associated with the document." } ], "returns": "Promise<void>", "description": "Embed text and store in the vector store." }, { "name": "search", "params": [ { "name": "query", "type": "string", "description": "Search query text." }, { "name": "topK", "type": "number", "description": "Number of results to return." } ], "returns": "Promise<SearchResult[]>", "description": "Search for similar documents." }, { "name": "remove", "params": [ { "name": "id", "type": "string", "description": "Document ID to remove." } ], "returns": "void", "description": "Remove a document from the store." } ], "outputs": [ { "name": "searchResults", "type": "SearchResult[]", "fields": [ { "name": "id", "type": "string" }, { "name": "text", "type": "string" }, { "name": "score", "type": "number" }, { "name": "metadata", "type": "Record<string, any>" } ], "description": "Ranked results with similarity scores." } ], "runtime": "universal", "version": "0.1.0", "category": "ai", "languages": [ "typescript", "python" ], "description": "In-memory vector store for embeddings with cosine similarity search. Supports OpenAI and Ollama embedding providers. Ideal for RAG pipelines and semantic search.", "dependencies": { "radzor": [], "packages": {} }, "composability": { "connectsTo": [ { "output": "searchResults", "mapField": "text", "compatibleWith": [ "@radzor/llm-completion.action.complete.prompt" ] } ] } }

@radzor/embeddings-store

Install

Inputs

Outputs

Actions

Events

Composability

radzor.manifest.json

Version History