How can i change the dimensions of the embeddings using the embedding model: llama-text-embed-2 hosted by pinecone

Support
,
1

this is my function and when i pass the dimension in parameters as 2048 it still create embedding of 1024, in docs i cant see where to pass dimension parameter, so i have kept it in parameters

async def embeddings(inputs: list, embedding_model: str = "llama-text-embed-v2" , input_type:str = "passage", truncate:str = "END", dimension: int = 2048):
    payload = {
        "model": embedding_model,
        "parameters": {
            "input_type": input_type,
            "truncate": truncate,
            "dimension": dimension
        },
        "inputs": inputs
    }
    
    headers = {
        "Api-Key": os.getenv("PINECONE_API_KEY"),
        "Content-Type": "application/json",
        "X-Pinecone-API-Version": "2025-01"
    }
    
    url = "https://api.pinecone.io/embed"
    
    async with httpx.AsyncClient() as client:
        response = await client.post(url, headers=headers, json=payload)
        response.raise_for_status()
        print("embeddings generated")
        return response.json()
2

@jay.pce21
Can you try it with the ‘2025-04’ api version instead of ‘2025-01’ in the “X-Pinecone-API-Version” header? It should work with this version & beyond of the API. We will add some documentation around this.

2 Likes
3

yeah it worked, thanks!!!