Embeddings Endpoint

Objective: Generate embeddings for the text "encode this" using the model text-embedding-ada-002 via 2 methods:

1. Konko AI's Python SDK
2. HTTP

No prior experience with Konko AI? No worries. This tutorial requires no preliminary knowledge and can be completed in less than 5 minutes.

🚧

Prerequisite: To follow along, you'll need a Konko API key.

Get one at https://www.konko.ai/.

Method #1: Python SDK

Step #1: Set Up & Install SDK
If you haven't already, first check out the Setup Access & SDK.

Step #2: Initialize & Set API Keys
After obtaining your Konko and OpenAI API keys, initialize them with the following commands:

import konko

konko.set_api_key('YOUR_KONKO_API_KEY')
konko.set_openai_api_key('YOUR_OPENAI_API_KEY')

Step #3: Generate Embeddings
To get the embeddings for your desired text, use the following commands:

response = konko.Embedding.create(
    model="text-embedding-ada-002",
    input="encode this"
)

This will return an array of floating point numbers, representing the embedding for the input text. Here's a truncated example of what the output might look like:

{
  "object": "list",
  "data": [
    {
      "object": "embedding",
      "index": 0,
      "embedding": [
        ...................
        -0.008713914,
        0.01852663,
        ... (and so on)
      ]
    }
  ],
  "model": "text-embedding-ada-002-v2",
  "usage": {
    "prompt_tokens": 2,
    "total_tokens": 2
  }
}

Method #2: HTTP

You can recreate the above steps in any major programming language. Simply refer to Konko's interactive API Reference docs to understand the HTTP endpoints and how to call them.