main.py

import os
from fastapi import FastAPI, Request, HTTPException
from sentence_transformers import SentenceTransformer

os.environ["OMP_NUM_THREADS"] = "2"
os.environ["MKL_NUM_THREADS"] = "2"

app = FastAPI(title="OpenAI Compatible API")

# Keep the model you specifically chose to host
MODEL_ID = "Snowflake/snowflake-arctic-embed-l"
model = SentenceTransformer(MODEL_ID, device="cpu")

@app.get("/health")
def health_check():
    return {"status": "healthy"}

@app.post("/v1/embeddings")
async def create_embeddings(request: Request):
    try:
        data = await request.json()
    except Exception:
        raise HTTPException(status_code=400, detail="Invalid JSON format")
    
    inputs = data.get("input")
    
    # 🛡️ THE SAFEGUARD: If n8n sends an empty string, force a valid text 
    # so the model generates real numbers instead of an array of zeros.
    if not inputs or inputs == [""] or inputs == "":
        inputs = ["dummy text to prevent zero vector database crash"]
        
    if isinstance(inputs, str):
        inputs = [inputs]

    # Generate vectors
    embeddings = model.encode(inputs, normalize_embeddings=True).tolist()
    
    response_data = [
        {
            "object": "embedding",
            "embedding": emb,
            "index": i
        }
        for i, emb in enumerate(embeddings)
    ]
    
    return {
        "object": "list",
        "data": response_data,
        "model": MODEL_ID,
        "usage": {"prompt_tokens": 0, "total_tokens": 0}
    }