Retrieval

GenAIScript provides various utilities to retrieve content and augment the prompt. This technique is typically referred to as RAG (Retrieval-Augmentation-Generation) in the literature.

Vector Search

GenAIScript provides various vector database to support embeddings search and retrieval augmented generation (RAG).

The retrieve.index creates or loads an existing index.

// index creation
const index = await retrieval.index("animals")
// indexing
await index.insertOrUpdate(env.files)
// search
const res = await index.search("cat dog")
def("RAG", res)

Local Index

By default, vector are stored locally in files under the .genaiscript folder using a local vector database based on vectra. The embeddings are computed using the embeddings model alias (it can also be configured through the options).

const index = await retrieval.index("animals", {
    embeddingsModel: "ollama:nomic-embed-text",
})

The index is serialized by default. If you wish to reset it on every execution, set deleteIfExists: true.

Azure AI Search

GenAIScript also supports using an Azure AI Search service. The Azure AI Search uses the simple query syntax.

const index = retrieval.index("animals", { type: "azure_ai_search" })

To configure the service, you will need to set the AZURE_AI_SEARCH_ENDPOINT and AZURE_AI_SEARCH_API_KEY environment variables in your .env file. Please refer to the Authentication documentation for more details.

AZURE_AI_SEARCH_ENDPOINT=https://{{service-name}}.search.windows.net/
AZURE_AI_SEARCH_API_KEY=...

Further index management can be done through the Azure Portal.

Fuzz Search

The retrieve.fuzzSearch performs a “traditional” fuzzy search to find the most similar documents to the prompt.

const files = await retrieval.fuzzSearch("cat dog", env.files)

Web Search

The retrieval.webSearch performs a web search using a search engine API. You will need to provide API keys for the search engine you want to use.

const { webPages } = await retrieval.webSearch("cat dog")
def("RAG", webPages)

Bing

To enable Bing search, configure the BING_SEARCH_API_KEY secret in your .env file. Learn more about configuring the Bing Search API.