Skip to main content

Gravix Layer x Pinecone Integration

Combine Gravix Layer's LLMs with Pinecone for scalable, high-performance vector search and retrieval-augmented generation (RAG).

What You'll Learn​

  • How to use Gravix Layer for embedding generation
  • How to store and search embeddings in Pinecone
  • Example: RAG pipeline with Gravix Layer and Pinecone

1. Install Required Packages​

pip install pinecone-client openai langchain python-dotenv

2. Configure Your API Key​

Add your API key to a .env file:

GRAVIXLAYER_API_KEY=your_api_key_here

3. Using Pinecone with Gravix Layer​

import os
import json
from openai import OpenAI
from pinecone import Pinecone, ServerlessSpec
from dotenv import load_dotenv

load_dotenv()
api_key = os.environ.get("GRAVIXLAYER_API_KEY", "test_key")
pinecone_api_key = os.environ.get("PINECONE_API_KEY", "your_real_pinecone_key")

client = OpenAI(
    base_url="https://api.gravixlayer.com/v1/inference",
    api_key=api_key,
)

text = "Why is the sky blue?"
embedding_response = client.embeddings.create(
    model="meta-llama/llama-3.1-8b-instruct",
    input=text,
)
vector = embedding_response.data[0].embedding

pc = Pinecone(api_key=pinecone_api_key)
index_name = "my-index"

if index_name not in pc.list_indexes().names():
    pc.create_index(
        name=index_name,
        dimension=len(vector),
        metric="cosine",
        spec=ServerlessSpec(cloud="aws", region="us-west-2")
    )
index = pc.Index(index_name)
index.upsert([{"id": "id1", "values": vector}])

results = index.query(vector=vector, top_k=3)
print(json.dumps(results.to_dict(), indent=2))

Expected Output:

{
  "data": [
    {
      "embedding": [ ... list of floats ... ],
      "index": 0,
      "object": "embedding"
    }
  ],
  "model": "meta-llama/llama-3.1-8b-instruct",
  "object": "list",
  "usage": { "prompt_tokens": 6, "total_tokens": 6 }
}

Note: You must set a valid Pinecone API key in your environment for the workflow to succeed. The code is fully compatible with Gravix Layer and Pinecone.

Pinecone enables scalable vector search. Gravix Layer provides fast, accurate embeddings for any RAG or search application.