Implementation:Groq Groq python EmbeddingResponse Usage

Knowledge Sources	Groq Python SDK
Domains	NLP, Embeddings, Data_Parsing
Last Updated	2026-02-15 16:00 GMT

Overview

Concrete Pydantic models for parsing embedding response data in the Groq Python SDK.

Description

The CreateEmbeddingResponse class contains a data list of Embedding objects and a usage object. Each Embedding has an embedding field (Union[List[float], str] depending on encoding format), an index (position in input), and an object type.

Usage

Returned automatically by client.embeddings.create(). Access vectors via response.data[i].embedding and usage via response.usage.total_tokens.

Code Reference

Source Location

• Repository: groq-python
• File: src/groq/types/create_embedding_response.py (L1-33)
• File: src/groq/types/embedding.py (L1-25)

Signature

class CreateEmbeddingResponse(BaseModel):
    data: List[Embedding]
    model: str
    object: Literal["list"]
    usage: Usage

class Embedding(BaseModel):
    embedding: Union[List[float], str]
    index: int
    object: Literal["embedding"]

class Usage(BaseModel):
    prompt_tokens: int
    total_tokens: int

Import

from groq.types import CreateEmbeddingResponse
from groq.types import Embedding

I/O Contract

Inputs

Name	Type	Required	Description
(object)	CreateEmbeddingResponse	Yes	Returned from Embeddings.create()

Outputs

Name	Type	Description
data[i].embedding	Union[List[float], str]	The embedding vector (float list or base64 string)
data[i].index	int	Position index in input array
usage.prompt_tokens	int	Tokens consumed
usage.total_tokens	int	Total tokens used

Usage Examples

response = client.embeddings.create(
    input=["hello", "world"],
    model="nomic-embed-text-v1_5",
)

# Extract vectors
for item in response.data:
    vector = item.embedding  # List[float]
    print(f"Index {item.index}: {len(vector)} dimensions")

# Usage stats
print(f"Prompt tokens: {response.usage.prompt_tokens}")
print(f"Total tokens: {response.usage.total_tokens}")

Related Pages

Implements Principle

• Principle:Groq_Groq_python_Vector_Extraction

Requires Environment

• Environment:Groq_Groq_python_Python_Groq_SDK