Fix: Install llama-cpp-python at runtime (HF Spaces workaround)

Dockerfile

@@ -7,16 +7,10 @@ RUN apt-get update && apt-get install -y \
     gcc \
     g++ \
     make \
-    wget \
     && rm -rf /var/lib/apt/lists/*
 
-# Copy requirements and install
+# Copy requirements and install (llama-cpp-python installed at runtime)
 COPY requirements.txt .
-
-# Install llama-cpp-python with pre-built wheels (much faster)
-RUN pip install --no-cache-dir llama-cpp-python==0.3.14 --extra-index-url https://abetlen.github.io/llama-cpp-python/whl/cpu
-
-# Install remaining requirements
 RUN pip install --no-cache-dir -r requirements.txt
 
 # Copy application code

main.py

@@ -1,3 +1,10 @@
+# Install llama-cpp-python at runtime (HF Spaces workaround)
+import subprocess
+import sys
+print("🔧 Installing llama-cpp-python at runtime...")
+subprocess.run([sys.executable, "-m", "pip", "install", "-v", "llama-cpp-python==0.3.14"], check=True)
+print("✅ llama-cpp-python installed successfully")
+
 from fastapi import FastAPI, HTTPException, WebSocket, WebSocketDisconnect, BackgroundTasks
 from fastapi.responses import StreamingResponse, JSONResponse, FileResponse
 from fastapi.middleware.cors import CORSMiddleware