Embeddings

POST https://api.staik.se/v1/embeddings

OpenAI-compatible embedding endpoint with bge-m3 — multilingual (excellent at Swedish), 1024-dimensional vectors, 8,192 token context. Building RAG or semantic search? Start here.

from openai import OpenAI

client = OpenAI(api_key="sk-st-your-key", base_url="https://api.staik.se/v1")

response = client.embeddings.create(
    model="bge-m3:latest",
    input=["First document to index", "Second document"],
)
vectors = [d.embedding for d in response.data]  # 1024 dimensions per vector

input takes a string or a list of strings. The response follows the OpenAI format: data[].embedding (1024 floats), data[].index, and usage.total_tokens.

With pgvector

Configure the column for 1024 dimensions:

CREATE EXTENSION IF NOT EXISTS vector;

CREATE TABLE documents (
  id bigserial PRIMARY KEY,
  content text,
  embedding vector(1024)
);

-- Nearest neighbors via cosine distance
SELECT content FROM documents
ORDER BY embedding <=> $1
LIMIT 5;