diff --git a/docs/ramalama-bench.1.md b/docs/ramalama-bench.1.md
index 87d51acf..c94435f3 100644
--- a/docs/ramalama-bench.1.md
+++ b/docs/ramalama-bench.1.md
@@ -28,6 +28,9 @@ URL support means if a model is on a web site or even on your local system, you
 #### **--help**, **-h**
 show this help message and exit
 
+#### **--network-mode**=*none*
+set the network mode for the container
+
 ## DESCRIPTION
 Benchmark specified AI Model.
 
diff --git a/docs/ramalama-convert.1.md b/docs/ramalama-convert.1.md
index 19fad843..9feb7857 100644
--- a/docs/ramalama-convert.1.md
+++ b/docs/ramalama-convert.1.md
@@ -25,6 +25,9 @@ type of OCI Model Image to convert.
 | car  | Includes base image with the model stored in a /models subdir |
 | raw  | Only the model and a link file model.file to it stored at /   |
 
+#### **--network-mode**=*none*
+sets the configuration for network namespaces when handling RUN instructions
+
 ## EXAMPLE
 
 Generate an oci model out of an Ollama model.
diff --git a/docs/ramalama-run.1.md b/docs/ramalama-run.1.md
index e0a69bd6..b318e4a1 100644
--- a/docs/ramalama-run.1.md
+++ b/docs/ramalama-run.1.md
@@ -53,6 +53,9 @@ llama.cpp explains this as:
 #### **--tls-verify**=*true*
 require HTTPS and verify certificates when contacting OCI registries
 
+#### **--network-mode**=*none*
+set the network mode for the container
+
 ## DESCRIPTION
 Run specified AI Model as a chat bot. RamaLama pulls specified AI Model from
 registry if it does not exist in local storage. By default a prompt for a chat
diff --git a/docs/ramalama-serve.1.md b/docs/ramalama-serve.1.md
index 49840ef7..972e894b 100644
--- a/docs/ramalama-serve.1.md
+++ b/docs/ramalama-serve.1.md
@@ -64,6 +64,9 @@ IP address for llama.cpp to listen on.
 #### **--name**, **-n**
 Name of the container to run the Model in.
 
+#### **--network-mode**=*bridge*
+set the network mode for the container
+
 #### **--port**, **-p**
 port for AI Model server to listen on
 
diff --git a/ramalama/cli.py b/ramalama/cli.py
index 0b3577b4..828afbe2 100644
--- a/ramalama/cli.py
+++ b/ramalama/cli.py
@@ -379,6 +379,12 @@ def bench_cli(args):
 
 def bench_parser(subparsers):
     parser = subparsers.add_parser("bench", aliases=["benchmark"], help="benchmark specified AI Model")
+    parser.add_argument(
+        "--network-mode",
+        type=str,
+        default="none",
+        help="set the network mode for the container",
+    )
     parser.add_argument("MODEL")  # positional argument
     parser.set_defaults(func=bench_cli)
 
@@ -600,6 +606,13 @@ def convert_parser(subparsers):
 Model "car" includes base image with the model stored in a /models subdir.
 Model "raw" contains the model and a link file model.file to it stored at /.""",
     )
+    # https://docs.podman.io/en/latest/markdown/podman-build.1.html#network-mode-net
+    parser.add_argument(
+        "--network-mode",
+        type=str,
+        default="none",
+        help="sets the configuration for network namespaces when handling RUN instructions",
+    )
     parser.add_argument("SOURCE")  # positional argument
     parser.add_argument("TARGET")  # positional argument
     parser.set_defaults(func=convert_cli)
@@ -717,6 +730,15 @@ def _run(parser):
 def run_parser(subparsers):
     parser = subparsers.add_parser("run", help="run specified AI Model as a chatbot")
     _run(parser)
+    # Disable network access by default, and give the option to pass any supported network mode into
+    # podman if needed:
+    # https://docs.podman.io/en/latest/markdown/podman-run.1.html#network-mode-net
+    parser.add_argument(
+        "--network-mode",
+        type=str,
+        default="none",
+        help="set the network mode for the container",
+    )
     parser.add_argument("MODEL")  # positional argument
     parser.add_argument(
         "ARGS", nargs="*", help="Overrides the default prompt, and the output is returned without entering the chatbot"
@@ -742,6 +764,16 @@ def serve_parser(subparsers):
     parser.add_argument(
         "-p", "--port", default=config.get('port', "8080"), help="port for AI Model server to listen on"
     )
+    # --network-mode=bridge lets the container listen on localhost, and is an option that's compatible
+    # with podman and docker:
+    # https://docs.podman.io/en/latest/markdown/podman-run.1.html#network-mode-net
+    # https://docs.docker.com/engine/network/#drivers
+    parser.add_argument(
+        "--network-mode",
+        type=str,
+        default="bridge",
+        help="set the network mode for the container",
+    )
     parser.add_argument("MODEL")  # positional argument
     parser.set_defaults(func=serve_cli)
 
diff --git a/ramalama/model.py b/ramalama/model.py
index 710a27ac..6417090b 100644
--- a/ramalama/model.py
+++ b/ramalama/model.py
@@ -153,6 +153,7 @@ def setup_container(self, args):
             "-i",
             "--label",
             "RAMALAMA",
+            f"--network={args.network_mode}",
             "--security-opt=label=disable",
             "--name",
             name,
diff --git a/ramalama/oci.py b/ramalama/oci.py
index 6f5e50ff..0d705174 100644
--- a/ramalama/oci.py
+++ b/ramalama/oci.py
@@ -174,7 +174,7 @@ def build(self, source, target, args):
             else:
                 c.write(model_raw)
         imageid = (
-            run_cmd([self.conman, "build", "--no-cache", "-q", "-f", containerfile.name, contextdir], debug=args.debug)
+            run_cmd([self.conman, "build", "--no-cache", f"--network={args.network_mode}", "-q", "-f", containerfile.name, contextdir], debug=args.debug)
             .stdout.decode("utf-8")
             .strip()
         )