Ollama - Install and Use OpenClaw (Local LLMs)

INDEX

• OpenClaw - Pre-Requisites
• OpenClaw - Install & Configure
• OpenClaw & Ollama - Basic CMDs
• OpenClaw - Models

OpenClaw - Pre-Requisites

To install:

• nodejs
• npm

On the Ubuntu machine itself, run:

sudo apt update
sudo apt install -y nodejs npm

Or:

# 1) Install a supported Node.js version
curl -fsSL https://deb.nodesource.com/setup_22.x | sudo -E bash -
sudo apt install -y nodejs

# 2) Verify the version is new enough
node -v
npm -v

Fix it with a user-owned npm prefix:

mkdir -p "$HOME/.local"
npm config set prefix "$HOME/.local"
echo 'export PATH="$HOME/.local/bin:$PATH"' >> ~/.bashrc
source ~/.bashrc

npm i -g openclaw@latest

OpenClaw - Install & Configure

QUICK START

ollama launch openclaw

Telegram configuration:

openclaw config set channels.telegram.dmPolicy "allowlist"
openclaw config set channels.telegram.allowFrom '["YOUR_USER_ID"]'

To find your Telegram user ID, use any of these:

• Ask the Telegram bot @userinfobot
• Use a Telegram API tool
• Check your Telegram app details / account info

openclaw config set channels.telegram.dmPolicy "allowlist"
openclaw config set channels.telegram.allowFrom '["123456789"]'

OpenClaw requires a larger context window.
It is recommended to use a context window of at least 64k tokens if using local models.

WEB SEARCH AND FETCH

OpenClaw ships with a bundled Ollama web_search provider that lets local or cloud-backed Ollama setups search the web through the configured Ollama host.

ollama launch openclaw

Ollama web search is enabled automatically when launching OpenClaw through Ollama.

To configure it manually:

openclaw configure --section web

Ollama web search for local models requires ollama signin.

Or just use DuckDuckGo, which is free and does not require sign-in.

CONFIGURE WITHOUT LAUNCHING

To change the model without starting the gateway and TUI:

ollama launch openclaw --config

To use a specific model directly:

ollama launch openclaw --model kimi-k2.5:cloud

If the gateway is already running, it restarts automatically to pick up the new model.

RECOMMENDED MODELS

Cloud models:

• kimi-k2.5:cloud — Multimodal reasoning with subagents
• qwen3.5:cloud — Reasoning, coding, and agentic tool use with vision
• glm-5.1:cloud — Reasoning and code generation
• minimax-m2.7:cloud — Fast, efficient coding and real-world productivity

Local models:

• gemma4 — Reasoning and code generation locally (~16 GB VRAM)
• qwen3.5 — Reasoning, coding, and visual understanding locally (~11 GB VRAM)

More models at ollama.com/search.

NON-INTERACTIVE (HEADLESS) MODE

Run OpenClaw without interaction for use in Docker, CI/CD, or scripts:

ollama launch openclaw --model kimi-k2.5:cloud --yes

The --yes flag auto-pulls the model, skips selectors, and requires --model to be specified.

CONNECT MESSAGING APPS

openclaw configure --section channels

Link WhatsApp, Telegram, Slack, Discord, or iMessage to chat with your local models from anywhere.

STOPPING THE GATEWAY

openclaw gateway stop

OpenClaw & Ollama - Basic CMDs

Base: https://docs.ollama.com/integrations/openclaw

OLLAMA

START

ollama launch openclaw
ollama launch openclaw --config

INSTALL

ollama pull <model_name>

RUN (INSTALL IF NOT PRESENT)

ollama pull <model_name>

OPENCLAW

openclaw onboard                     # Setup Wizard
openclaw configure                   # Specific config
openclaw configure --section web

openclaw gateway start              # Start a managed gateway
openclaw gateway run                # Run the gateway in the foreground

openclaw                            # TUI: Chat in CLI
openclaw dashboard                  # Get HTTP link

STATUS

openclaw status --all
openclaw gateway status --deep     # For details
openclaw memory status --deep

HOOKS

openclaw hooks list               # "ready" means enabled
openclaw hooks enable <name>
openclaw hooks disable <name>

SKILLS

openclaw skills list
openclaw skills enable <name>
openclaw skills disable <name>

openclaw skills search <xxx>      # ClawHub-backed skills
openclaw skills install <xxx>     # ClawHub-backed skills
openclaw skills update <xxx>      # ClawHub-backed skills

FIX

openclaw doctor
openclaw doctor --fix
openclaw security audit --deep

OpenClaw - Models

Model Benchmark Sources

• BenchLM: https://benchlm.ai/
• SWE-Bench: https://www.swebench.com/
• LiveCodeBench: https://livecodebench.github.io/leaderboard.html

VRAM - Rules of Thumb

• Q4_K_M size estimate: ~0.6 GB per billion parameters
• Always leave 20-30% headroom for context (KV cache)
• Example: A 24GB card comfortably runs a ~16-19GB model
• It will NOT run a ~21GB+ model with useful context lengths

Windows 11 - Local Account

If you dont want to put your email on windows and create a local account like in previous wondows version just follow this:

 - Install windows 11 up to this screen   

   "Is this the rigth country or region ?"

 - Shift + F10

 - Command line (CLI) will show 

 - Click a bit on the Window

 - Type on CLI 

   > oobe\bypassnro

 

 - The machine will reboot and will get

   back to up to this screen:

   "Is this the rigth country or region ?"

 - Shift + F10

 - Command line (CLI) will show  

 - Type on CLI:

   > ipconfig /release

   disconnects the machin from the network

  - Close CLI

 

- Continue... Next... 

 - Install windows 11 up to this screen:

   "Let´s connect you to a network"

   select:

   "I dont have Internet"

   then:

   "Continue with limited setup"

 - And it will go the local account setup

 

 - Now its business as usual... Next... Next...

NVIDIA TESLA P100 - Windows and VMWare (Software Versions)

 The NVIDIA Tesla P100 is a server GPU, thas has become afordable recently, you can use it on Windows directly or as a GRID GPU on VMWARE ESXi (shared between VMs). 

As this is an older GPU it's not supported on the latest drivers in Windows or the latest VMWARE ESXi (it's well supported on Proxmox).

Windows

You can get the windows drivers (Grid vGPU) here:

https://cloud.google.com/compute/docs/gpus/grid-drivers-table?hl=pt-br#windows_drivers

the latest driver that will support the P100 is:

        538.33

from example version 551.61 does not support the P100:

VMware ESXi

The latest version of ESXi that suports the P100 as a Grid GPU is ESXi 6.7.

On later versions of VMWare you can only do Passthrough to one VM.

Hope this helps someone to avoid losing as much time as I did trying finding the rigth versions of software that support the NVIDIA Tesla P100.

VMWare ESXi (VSphere) - How to Downgrade VM Hardware (example 7U3 to 6.7)

Upgrading the hardware on a VM is easy just rigth click on the mouse and chose "Upgrade VM Compatibility".

The inverse, which downgrading VM Hardware, there is no official way of doing it.

To downgrade from ESXI 7U3 to 6.7 I did it like this (should work for other versions):

• Enable SSH on ESXI7

• WinSCP to ESXI7

• Clone the .vmx to save it before changing it

• Change original .vmx to ESXI 6.7 (hardware version 14):

            virtualHW.version = "14"

            You can find other harware versions at:

            https://knowledge.broadcom.com/external/article?legacyId=1003746

                VMware products and their virtual hardware version

• Unregister the VM

• Register VM using the modified .vmx



Ansible - Setup on Linux ( +cisco devices )

 

Ansible Installation

On the ubuntu shell run:

sudo apt update
sudo apt upgrade -y

 

# PYTHON

sudo apt-get install python3

sudo apt install python3-pip

 

# ANSIBLE

sudo apt install -y software-properties-common

sudo apt install -y ansible

 

# ANSIBLE COMMON MODULES / LIBRARIES (OPTIONAL)

ansible-galaxy collection install cisco.ios

sudo apt-get install python3-dnspython

sudo pip install ansible-pylibssh

sudo pip install pexpect

sudo pip install paramiko

sudo apt-get install sshpass

SSH Host Key Checking

If you want to avoid issues on connecting via SSH because the the keys are not added, go to:  

sudo nano /etc/ansible/ansible.cfg

and uncoment the "host_key_checking" line:

[defaults]
host_key_checking = False

Verification

WSL

Win+r ubuntu

 

PYTHON

python3

 

ANSIBLE

ansible-playbook --version

Ansible - Setup on Windows WSL ( +cisco devices )

Enable Windows Features for WSL (Windows Subsistem for Linux - aka Linux on Windows)

Run on powershell

dism.exe /online /enable-feature /featurename:Microsoft-Windows-Subsystem-Linux /all /norestart

dism.exe /online /enable-feature /featurename:VirtualMachinePlatform /all /norestart

 

Download and Install WSL2 Patch

https://wslstorestorage.blob.core.windows.net/wslblob/wsl_update_x64.msi

Set Default Version to Version 2

Run on powershell

wsl --set-default-version 2

Ubuntu WSL Download - Without Using Windows Store - Online Method

Open windows CMD and run:

wsl --list --online
The following is a list of valid distributions that can be installed.
Install using 'wsl --install -d <Distro>'.

NAME FRIENDLY NAME
Ubuntu Ubuntu
Debian Debian GNU/Linux
kali-linux Kali Linux Rolling
openSUSE-42 openSUSE Leap 42
SLES-12 SUSE Linux Enterprise Server v12
Ubuntu-16.04 Ubuntu 16.04 LTS
Ubuntu-18.04 Ubuntu 18.04 LTS
Ubuntu-20.04 Ubuntu 20.04 LTS

wsl --install --distribution Ubuntu

Ubuntu WSL Download - Without Using Windows Store - Offline Method

Open windows powershell and run:

Invoke-WebRequest -Uri https://aka.ms/wslubuntu2004 -OutFile Ubuntu.appx

Add-AppxPackage .\Ubuntu.appx

First Run and Installation

Win+r Ubuntu

Set WSL DNS to Google DNS (default is a Windows generated, which doesn't always work)

Change the DNS to Google's DNS

sudo nano /etc/resolv.conf

# This file was automatically generated by WSL. To stop automatic generation of this file, add the following entry to /etc/wsl.conf:
# [network]
# generateResolvConf = false
nameserver 8.8.8.8

Avoid windows from changing the DNS server again:

sudo nano /etc/wsl.conf 

[network]
generateResolvConf = false

Ansible Installation

On the ubuntu shell run:

sudo apt update
sudo apt upgrade -y

 

# PYTHON

sudo apt-get install python3

sudo apt install python3-pip

 

# ANSIBLE

sudo apt install -y software-properties-common

sudo apt install -y ansible

 

# ANSIBLE COMMON MODULES / LIBRARIES (OPTIONAL)

ansible-galaxy collection install cisco.ios

sudo apt-get install python3-dnspython

sudo pip install ansible-pylibssh

sudo pip install pexpect

sudo pip install paramiko

sudo apt-get install sshpass

SSH Host Key Checking

If you want to avoid issues on connecting via SSH because the the keys are not added, go to:  

sudo nano /etc/ansible/ansible.cfg

and uncoment the "host_key_checking" line:

[defaults]
host_key_checking = False

Extra - Convert Install Machines to WSL2

If you already had WSL installed and some distros and want to convert then, first check your distros and versions:

wsl --list -v
NAME STATE VERSION
* Ubuntu Running 1

then change it to version two and windows will start the conversion:

wsl --set-version Ubuntu 2

Verification

WSL

Win+r ubuntu

 

PYTHON

python3

 

ANSIBLE

ansible-playbook --version

Nvidia Shield - NAS Server (SMBv1)

Configuring The Nvidia Shield (v8.0.0)

Go to Settings > Storage & reset > Transfer files over network, to turn this feature on.

You'll be presented with a

• User name - the Nvidia Shield generates it but you can change it
• Password - the Nvidia Shield generates the password not you

that you can use on a networked PC to access SHIELD folders.

Make sure your PC is on the same network as your SHIELD.

Activate SMB1 Client on Windows 10

This is a legacy protocol and it´s client is disabled by default, to enable it:

1. Open Control Panel.
2. Click on Programs.
3. Click on Turn Windows features on or off link.

4. Expand the SMB 1.0/CIFS File Sharing Support option.
5. Check the SMB 1.0/CIFS Client option.

6. Click the OK button.
7. Click the Restart now button.

After completing these steps, you'll once again be able to see and connect to network devices running the old protocol on your local network from your Windows 10 computer.

Accessing The Nvidia Shield Shared Folder

On windows start menu run the following command:

You can also permanently map your SHIELD as a network drive by right-clicking on "Network" in File Explorer and selecting "Map Network Drive". Select an unused Drive letter.

Based On:

https://nvidia.custhelp.com/app/answers/detail/a_id/4172/~/how-do-i-share-shield-tv-folders-with-a-pc