README: clean up

README.md

@@ -34,12 +34,16 @@ curl -fsSL https://pkg.gpud.dev/install.sh | sh
 
 Note that the install script doesn't support other architectures (arm64) and OSes (macos), yet.
 
-### Run locally (self-hosted option)
+### Run GPUd with Lepton Platform
 
-For linux, run the following command to start the service (self-hosted option):
+Sign up at [lepton.ai](https://www.lepton.ai/) and get the workspace token from the "Settings" and "Tokens" page:
+
+<img src="./assets/gpud-lepton.ai-machines-settings.png" width="80%" alt="GPUd lepton.ai machines settings">
+
+Copy the token in the format of `workspace:token` and pass it to the `gpud up --token` flag:
 
 ```bash
-sudo gpud up
+sudo gpud up --token <LEPTON_AI_WORKSPACE:TOKEM>
 ```
 
 To check the status of the running gpud:
@@ -54,61 +58,37 @@ To check the logs of the running gpud:
 sudo gpud logs
 ```
 
-To access the local web UI, open https://localhost:15132 in your browser, as below:
-
-<img src="./assets/gpud-local-web-ui-1.png" width="80%" alt="GPUd local web UI 1">
-<img src="./assets/gpud-local-web-ui-2.png" width="80%" alt="GPUd local web UI 2">
+### Run GPUd standalone
 
-To disable the local web UI, pass the `--web-disable` flag in the following file:
+For linux, run the following command to start the service:
 
 ```bash
-vi /etc/default/gpud
+sudo gpud up
 ```
 
-```bash
-# gpud environment variables are set here
-FLAGS="--log-level=info --web-disable"
-```
+You can also start with the standalone mode and later switch to the managed option:
 
 ```bash
-sudo systemctl daemon-reload
-sudo systemctl restart gpud
+# when the token is ready, run the following command
+sudo gpud login --token <LEPTON_AI_WORKSPACE:TOKEM>
 ```
 
-### Report to lepton.ai (managed option)
-
-Optionally you may register your machine with Lepton AI Platform -- the managed option brings several benefits:
-
-- Automated GPU health check and repair.
-- Centralized GPU metrics and logs.
-- Real-time GPU failure detection and alerting.
+To access the local web UI, open https://localhost:15132 in your browser.
 
-> Please ensure that your machine has a public IP address and that the GPUd port (default 15132) is reachable.
-
-Sign up at [lepton.ai](https://www.lepton.ai/) and get the workspace token from the "Settings" and "Tokens" page:
-
-<img src="./assets/gpud-lepton.ai-machines-settings.png" width="80%" alt="GPUd lepton.ai machines settings">
-
-Copy the token in the format of `workspace:token` and pass it to the `gpud up --token` flag:
+To disable the local web UI, pass the `--web-disable` flag in the following file:
 
 ```bash
-sudo gpud up --token <LEPTON_AI_WORKSPACE:TOKEM>
+vi /etc/default/gpud
 ```
 
-Then see the "Machines" page to check the status of the machine:
-
-<img src="./assets/gpud-lepton.ai-machines-view.png" width="80%" alt="GPUd lepton.ai machines view">
-
-*The machine identifier is currently auto-generated.*
-
-You can also start with the self-hosted option and later switch to the managed option:
-
 ```bash
-# start without token
-sudo gpud up
+# gpud environment variables are set here
+FLAGS="--log-level=info --web-disable"
+```
 
-# when the token is ready, run the following command
-sudo gpud login --token <LEPTON_AI_WORKSPACE:TOKEM>
+```bash
+sudo systemctl daemon-reload
+sudo systemctl restart gpud
 ```
 
 #### If your system doesn't have systemd
@@ -131,10 +111,6 @@ It is possible that GPUd sends basic host information to lepton.ai to help under
 
 Once you opt-in to the lepton.ai platform, the GPUd periodically sends more detailed information about the host (e.g., GPU model and metrics), via the secure channel.
 
-#### Does my machine need a public IP to report to lepton.ai?
-
-No. Once registered, the GPUd creates a secure channel to the lepton.ai platform for sending metrics information.
-
 ### Stop and uninstall
 
 ```bash