Upload convert_to_gguf.py with huggingface_hub

convert_to_gguf.py

@@ -59,9 +59,9 @@ nproc_result = subprocess.run(["nproc"], capture_output=True, text=True, check=T
 nproc = nproc_result.stdout.strip()
 print(f"Building llama.cpp with {nproc} cores using CMake...")
 
-# Use CMake to build
+# Use CMake to build (CPU only - CUDA not needed for conversion/quantization)
 os.makedirs("llama.cpp/build", exist_ok=True)
-subprocess.run(["cmake", "-B", "llama.cpp/build", "-S", "llama.cpp", "-DGGML_CUDA=ON"], check=True)
+subprocess.run(["cmake", "-B", "llama.cpp/build", "-S", "llama.cpp"], check=True)
 subprocess.run(["cmake", "--build", "llama.cpp/build", "--config", "Release", "-j", nproc], check=True)
 
 # Step 5: Convert to GGUF format