Skip to main content
The Embeddings endpoint converts text into dense numerical vectors that capture semantic meaning. You can use these vectors to power semantic search, build retrieval-augmented generation (RAG) pipelines, cluster similar content, or compute similarity scores between pieces of text. NexLLM’s embeddings endpoint is fully OpenAI-compatible, so any existing embeddings workflow transfers without code changes.

​
Endpoint

POST https://www.nexllm.ai/v1/embeddings

​
Request Parameters

​
model
string
required
The embedding model to use. Example: text-embedding-3-small, text-embedding-3-large, text-embedding-ada-002.
​
input
string | array
required
The text to embed. Pass a single string for one embedding, or an array of strings to embed multiple texts in a single request. Batching multiple inputs is more efficient than making separate requests.

​
Response Fields

​
data
array
An array of embedding objects, one for each input string. The order matches the order of the inputs you provided.
​
data[].embedding
array of floats
The embedding vector for this input — an array of floating-point numbers. The dimensionality depends on the model you chose.
​
data[].index
integer
The zero-based index of this embedding in the data array, corresponding to the position of the input string you submitted.
​
usage.prompt_tokens
integer
The total number of tokens processed across all inputs in the request.

​
Code Example

from openai import OpenAI

client = OpenAI(
    api_key="sk-xxxxxxxxxxxxxxxx",
    base_url="https://www.nexllm.ai/v1"
)

response = client.embeddings.create(
    model="text-embedding-3-small",
    input="The quick brown fox jumps over the lazy dog"
)

print(response.data[0].embedding)
To embed multiple strings at once, pass a list as input: 
response = client.embeddings.create(
    model="text-embedding-3-small",
    input=[
        "The quick brown fox jumps over the lazy dog",
        "Pack my box with five dozen liquor jugs"
    ]
)

for item in response.data:
    print(f"Index {item.index}: {len(item.embedding)}-dimension vector")
When building a RAG pipeline, embed your document chunks once at ingestion time and store the vectors in a vector database (e.g. Pinecone, Weaviate, or pgvector). At query time, embed the user’s question and perform a nearest-neighbour search to retrieve relevant chunks before calling the Chat Completions endpoint.