embedbase

Project Introduction: Embedbase

Embedbase is an innovative solution designed to simplify the process of using cutting-edge technologies such as VectorDBs and Large Language Models (LLMs). With Embedbase, developers can easily access these advanced tools through a straightforward API, eliminating the need to host them independently.

What is Embedbase?

Embedbase provides a user-friendly API that makes it easy to leverage the power of Vector Databases and LLMs. This allows developers to integrate high-level applications without dealing with the complexities of hosting and managing these technologies.

Key Features of Embedbase

• Text Generation: Embedbase's .generateText() function supports the utilization of over nine different LLMs, making it versatile for different text generation needs.

• Semantic Search: With the .add() and .search() functions, users can create and query a semantically searchable dataset. This feature significantly enhances the ability to retrieve relevant information based on user queries.

Getting Started with Embedbase

To start using Embedbase, you can install it through npm with the command npm i embedbase-js. The following example demonstrates how to initialize a client, perform a semantic search, and use a language model to generate text:

import { createClient } from 'embedbase-js'
// initialize client
const embedbase = createClient(
  'https://api.embedbase.xyz',
  '<grab me here https://app.embedbase.xyz/>'
)

const question =
  'im looking for a nice pant that is comfortable and i can both use for work and for climbing'

// search for information in a pre-defined dataset and returns the most relevant data
const searchResults = await embedbase.dataset('product-ads').search(question)

// transform the results into a string so they can be easily used inside a prompt
const stringifiedSearchResults = searchResults
  .map(result => result.data)
  .join('')

const answer = await embedbase
  .useModel('openai/gpt-3.5-turbo')
  .generateText(`${stringifiedSearchResults} ${question}`)

console.log(answer) // 'I suggest considering harem pants for your needs. Harem pants are known for their ...'

Practical Applications

Embedbase is being leveraged in various innovative ways:

• Recommendation Engines: AVA integrates Embedbase to help users find related notes efficiently.
• Data Interactions: Solpilot uses Embedbase for smart contract integration, offering an auto-pilot feature.
• Documentation Search: Embedbase powers ChatGPT-driven search tools for markdown documentation, enhancing accessibility.

Explore Embedbase

To dive deeper, users can explore tutorials and documentation available on the Embedbase website. The documentation is supported by GPT-4, allowing interactive queries directly within the platform.

For those looking to join a community or seek support, the Embedbase Discord community is an invaluable resource where users can ask questions and engage with other members.

Becoming a Contributor

Embedbase encourages contributions! Interested parties can find detailed guidance on the contributing process in the CONTRIBUTING.md file. This contains information about code conduct and submitting pull requests.

The journey with Embedbase begins with a simple sign-up for the Embedbase Cloud, offering a free platform to get started with AI development. As projects grow and integrate more advanced AI functionalities, Embedbase provides a robust and easy-to-use suite of tools to enable developers to optimize their workflows.