diff --git a/unsloth/save.py b/unsloth/save.py
index ceb36854d2..714df7682a 100644
--- a/unsloth/save.py
+++ b/unsloth/save.py
@@ -131,6 +131,13 @@ ALLOWED_QUANTS = {
 }
 
 
+def has_curl():
+    return shutil.which("curl") is not None
+
+
+CURL_FLAG = "-DLLAMA_CURL=ON" if has_curl() else "-DLLAMA_CURL=OFF"
+
+
 def print_quantization_methods():
     for key, value in ALLOWED_QUANTS.items():
         print(f'"{key}"  ==> {value}')
@@ -879,8 +886,9 @@ def install_llama_cpp_make_non_blocking():
         # Uses new CMAKE
         n_jobs = max(int(psutil.cpu_count()), 1)  # Use less CPUs since 1.5x faster
         check = os.system(
-            "cmake llama.cpp -B llama.cpp/build -DBUILD_SHARED_LIBS=OFF -DGGML_CUDA=OFF -DLLAMA_CURL=ON"
+            f"cmake llama.cpp -B llama.cpp/build -DBUILD_SHARED_LIBS=OFF -DGGML_CUDA=OFF {CURL_FLAG}"
         )
+
         if check != 0:
             raise RuntimeError(
                 f"*** Unsloth: Failed compiling llama.cpp using os.system(...) with error {check}. Please report this ASAP!"
@@ -991,11 +999,12 @@ def install_llama_cpp_old(version = -10):
     if try_execute(commands) == "CMAKE":
         # Instead use CMAKE
         commands = [
-            "cmake llama.cpp -B llama.cpp/build -DBUILD_SHARED_LIBS=OFF -DGGML_CUDA=OFF -DLLAMA_CURL=ON",
+            f"cmake llama.cpp -B llama.cpp/build -DBUILD_SHARED_LIBS=OFF -DGGML_CUDA=OFF {CURL_FLAG}",
             f"cmake --build llama.cpp/build --config Release -j{psutil.cpu_count()*2} --clean-first --target {' '.join(LLAMA_CPP_TARGETS)}",
             "cp llama.cpp/build/bin/llama-* llama.cpp",
             "rm -rf llama.cpp/build",
         ]
+
         try_execute(commands)
 
     # Check if successful
@@ -1037,7 +1046,7 @@ def install_llama_cpp_blocking(use_cuda = False):
     if try_execute(commands) == "CMAKE":
         # Instead use CMAKE
         commands = [
-            "cmake llama.cpp -B llama.cpp/build -DBUILD_SHARED_LIBS=OFF -DGGML_CUDA=OFF -DLLAMA_CURL=ON",
+            f"cmake llama.cpp -B llama.cpp/build -DBUILD_SHARED_LIBS=OFF -DGGML_CUDA=OFF {CURL_FLAG}",
             f"cmake --build llama.cpp/build --config Release -j{psutil.cpu_count()*2} --clean-first --target {' '.join(LLAMA_CPP_TARGETS)}",
             "cp llama.cpp/build/bin/llama-* llama.cpp",
             "rm -rf llama.cpp/build",