Initial deploy of Ibani Inference API

Dockerfile

@@ -0,0 +1,15 @@
+FROM python:3.9-slim
+
+WORKDIR /app
+
+COPY requirements.txt .
+RUN pip install --no-cache-dir -r requirements.txt
+
+COPY . .
+
+# Expose the port FastAPI runs on
+EXPOSE 7860
+
+# Command to run the FastAPI app
+# Note: HF Spaces uses port 7860 by default
+CMD ["uvicorn", "app:app", "--host", "0.0.0.0", "--port", "7860"]

app.py

@@ -0,0 +1,45 @@
+from fastapi import FastAPI, HTTPException
+from pydantic import BaseModel
+from transformers import pipeline
+import torch
+import os
+
+app = FastAPI(title="Ibani Translator API")
+
+# Model configuration
+MODEL_ID = "williampepple1/ibani-translator"
+
+print(f"Loading model {MODEL_ID}...")
+try:
+    # Use pipeline for easy inference
+    # device=-1 forces CPU usage which is what HF Spaces free tier provides
+    translator = pipeline(
+        "translation", 
+        model=MODEL_ID, 
+        device=-1
+    )
+    print("Model loaded successfully!")
+except Exception as e:
+    print(f"Error loading model: {e}")
+    translator = None
+
+class TranslationRequest(BaseModel):
+    text: str
+
+@app.get("/")
+def read_root():
+    return {"status": "healthy", "model": MODEL_ID}
+
+@app.post("/translate")
+async def translate(request: TranslationRequest):
+    if translator is None:
+        raise HTTPException(status_code=503, detail="Model not loaded")
+    
+    try:
+        result = translator(request.text)
+        return {
+            "translated_text": result[0]['translation_text'],
+            "original_text": request.text
+        }
+    except Exception as e:
+        raise HTTPException(status_code=500, detail=str(e))

requirements.txt

@@ -0,0 +1,7 @@
+fastapi
+uvicorn
+transformers
+torch
+sentencepiece
+sacremoses
+pydantic