Upload app.py with huggingface_hub

app.py

@@ -0,0 +1,48 @@
+"""Embedding Server (sentence-transformers) for HuggingFace Spaces."""
+import os
+from fastapi import FastAPI
+from pydantic import BaseModel
+from sentence_transformers import SentenceTransformer
+
+MODEL_NAME = os.environ.get("MODEL_NAME", "sentence-transformers/multi-qa-MiniLM-L6-cos-v1")
+print(f"[Embedding] Loading model: {MODEL_NAME}...", flush=True)
+model = SentenceTransformer(MODEL_NAME)
+print("[Embedding] Model loaded.", flush=True)
+
+app = FastAPI()
+
+
+class EmbedRequest(BaseModel):
+    text: str | list[str]
+    normalize: bool = True
+
+
+class BatchEmbedRequest(BaseModel):
+    texts: list[str]
+    normalize: bool = True
+
+
+@app.get("/health")
+def health():
+    return {"status": "ok", "model": MODEL_NAME}
+
+
+@app.post("/embed")
+def embed(req: EmbedRequest):
+    texts = [req.text] if isinstance(req.text, str) else req.text
+    embeddings = model.encode(texts, normalize_embeddings=req.normalize)
+    return {
+        "embeddings": embeddings.tolist(),
+        "model": MODEL_NAME,
+        "dimensions": embeddings.shape[1],
+    }
+
+
+@app.post("/embed_batch")
+def embed_batch(req: BatchEmbedRequest):
+    embeddings = model.encode(req.texts, normalize_embeddings=req.normalize)
+    return {
+        "embeddings": embeddings.tolist(),
+        "model": MODEL_NAME,
+        "dimensions": embeddings.shape[1],
+    }