### Setup Python Environment and Download Client

Source: https://docs.lambda.ai/education/large-language-models/serving-llama-3-1-405b

Creates a Python virtual environment, installs the OpenAI library, and downloads the vLLM example client script. Ensure you are on the worker node before running.

```bash
python3 -m venv llama-3.1
source llama-3.1/bin/activate
pip install -U openai
curl -o ${SHARED_DIR}/inference_test.py 'https://raw.githubusercontent.com/vllm-project/vllm/main/examples/openai_chat_completion_client.py'
```

--------------------------------

### Setup Python Environment and Install OpenAI Client

Source: https://docs.lambda.ai/public-cloud/1-click-clusters/serving-llama-3_1-405b

Create and activate a Python virtual environment, then install the OpenAI Python package. This is a prerequisite for running the chat completion client script.

```bash
python3 -m venv llama-3.1
source llama-3.1/bin/activate
pip install -U openai
```

--------------------------------

### Install Lambda Stack

Source: https://docs.lambda.ai/education/linux-usage/lambda-stack-and-recovery-images

Installs the latest Lambda Stack by downloading and executing the installation script. Ensure you have the necessary permissions.

```bash
wget -nv -O- https://lambda.ai/install-lambda-stack.sh | sh - 

```

--------------------------------

### Install uv, set up venv, and install vLLM

Source: https://docs.lambda.ai/education/large-language-models/deploying-nemotron-3-nano

Installs the uv package manager, creates a Python virtual environment, and installs the vLLM library with PyTorch backend support. This is a prerequisite for serving models with vLLM.

```bash
curl -LsSf https://astral.sh/uv/install.sh | sh
uv venv --python 3.12 --seed
source .venv/bin/activate
uv pip install vllm --torch-backend=auto
```

--------------------------------

### Allocate Resources with salloc and Start Interactive Shell

Source: https://docs.lambda.ai/public-cloud/1-click-clusters/managed-slurm

Use salloc to request resources and start an interactive shell on the allocated node. This example requests one node with two GPUs.

```bash
salloc --gpus=2 --nodes=1 --ntasks-per-node=1 srun --pty /bin/bash

```

--------------------------------

### Install s5cmd

Source: https://docs.lambda.ai/public-cloud/importing-exporting-data

Download and install the s5cmd tool on your instance or node. Replace `<RELEASE-URL>` with the actual download link.

```bash
wget <RELEASE-URL>

```

```bash
sudo apt install ./<DEB-FILENAME>

```

--------------------------------

### Start VNC Server

Source: https://docs.lambda.ai/public-cloud/on-demand/connecting-instance

Starts the TightVNC server, prompting you to set a VNC access password.

```bash
vncserver
```

--------------------------------

### Create example directory

Source: https://docs.lambda.ai/education/scheduling-and-orchestration/orchestrating-workloads-with-dstack

Creates a new directory for dstack tutorial examples and navigates into it. This should be done in a separate terminal from the running dstack server.

```bash
mkdir ~/lambda-dstack-examples && cd ~/lambda-dstack-examples
```

--------------------------------

### Install Miniconda

Source: https://docs.lambda.ai/education/programming/virtual-environments-containers

Execute the downloaded Miniconda3 installer script. Follow the prompts to complete the installation, ensuring you allow initialization.

```bash
sh Miniconda3-latest-Linux-x86_64.sh
```

--------------------------------

### Install cryptsetup

Source: https://docs.lambda.ai/education/linux-usage/basic-linux-commands-and-system-administration

Installs the cryptsetup package, which is necessary for disk encryption.

```bash
sudo apt update && sudo apt -y install cryptsetup
```

--------------------------------

### Example: Get TensorFlow Build Info

Source: https://docs.lambda.ai/education/programming/virtual-environments-containers

Launches a TensorFlow NGC container and executes a Python command to retrieve detailed build information, including CUDA and cuDNN versions.

```bash
docker run --gpus all -it nvcr.io/nvidia/tensorflow:23.05-tf2-py3 python -c "import tensorflow as tf ; sys_details = tf.sysconfig.get_build_info() ; print(sys_details)"

```

--------------------------------

### Example Filesystem Usage Output

Source: https://docs.lambda.ai/public-cloud/filesystems

This is an example of the output from the `df -h -BG` command, illustrating how filesystem usage is displayed.

```text
Filesystem                                        1G-blocks  Used Available Use% Mounted on
tmpfs                                                   44G    1G       44G   1% /run
efivarfs                                                 1G    1G        1G   2% /sys/firmware/efi/efivars
/dev/vda1                                             3969G   24G     3946G   1% /
tmpfs                                                  216G    1G      216G   1% /dev/shm
tmpfs                                                    1G    0G        1G   0% /run/lock
/dev/vda15                                               1G    1G        1G   7% /boot/efi
10.12.69.11:/768f824a-2608-4200-8a6a-3d2996d10587  6792422G    0G  6792422G   0% /lambda/nfs/persistent-storage
tmpfs                                                   44G    1G       44G   1% /run/user/1000


```

--------------------------------

### Install curl and jq

Source: https://docs.lambda.ai/education/large-language-models/kubeai-hermes-3

Installs `curl` and `jq` on your instance. These tools are necessary for interacting with the KubeAI API.

```bash
sudo apt update && sudo apt -y install curl jq

```

--------------------------------

### Install dstack with Lambda support

Source: https://docs.lambda.ai/education/scheduling-and-orchestration/orchestrating-workloads-with-dstack

Installs the dstack library with specific support for Lambda Cloud. Ensure python3 and pip are installed.

```bash
pip install -U "dstack[lambda]"
```

--------------------------------

### Example User Directory Creation Command

Source: https://docs.lambda.ai/education/linux-usage/basic-linux-commands-and-system-administration

This is an example of the command to create a user-accessible directory named 'ubuntu' within the '/mnt/encrypted-drive' mount point.

```bash
sudo mkdir /mnt/encrypted-drive/ubuntu && \
    sudo chown $(id -u):$(id -g) /mnt/encrypted-drive/ubuntu
```

--------------------------------

### Install VNC Server and GNOME Desktop

Source: https://docs.lambda.ai/public-cloud/on-demand/connecting-instance

Installs the TightVNC server and GNOME desktop environment on your instance. Run this command after establishing an SSH connection.

```bash
sudo apt update &&
    sudo apt install -y tightvncserver gnome
```

--------------------------------

### Example Kubernetes Node Output

Source: https://docs.lambda.ai/public-cloud/1-click-clusters/managed-kubernetes/managed-kubernetes-legacy

Example output when listing nodes in an MK8s cluster.

```text
NAME        STATUS   ROLES                       AGE   VERSION
head-01     Ready    control-plane,etcd,master   8d    v1.32.3+rke2r1
head-02     Ready    control-plane,etcd,master   8d    v1.32.3+rke2r1
head-03     Ready    control-plane,etcd,master   8d    v1.32.3+rke2r1
worker-01   Ready    <none>                      8d    v1.32.3+rke2r1
worker-02   Ready    <none>                      8d    v1.32.3+rke2r1

```

--------------------------------

### Download and Install Guest Agent

Source: https://docs.lambda.ai/public-cloud/guest-agent

Download and execute the installation script for the Lambda Guest Agent. This command fetches the script and runs it with root privileges.

```bash
curl -L https://lambdalabs-guest-agent.s3.us-west-2.amazonaws.com/scripts/install.sh | sudo bash
```

--------------------------------

### Install System Packages

Source: https://docs.lambda.ai/education/scheduling-and-orchestration/skypilot-deploy-kubernetes

Installs necessary system packages including Python, pip, and utilities for managing the environment.

```bash
sudo apt update && sudo apt install -y python3 python3-venv python3-pip curl netcat socat

```

--------------------------------

### Start Ollama Server with Docker

Source: https://docs.lambda.ai/education/large-language-models/deepseek-r1-ollama

This command downloads the Ollama Docker image and starts the Ollama server, making it accessible via port 11434 and enabling GPU usage.

```bash
sudo docker run -d --gpus=all -v ollama:/root/.ollama -p 11434:11434 --name ollama ollama/ollama

```

--------------------------------

### Example rsync File Copy to Remote Server

Source: https://docs.lambda.ai/education/linux-usage/basic-linux-commands-and-system-administration

This example demonstrates copying a local directory 'rsync_example_dir' to the home directory of the 'ubuntu' user on a remote server.

```bash
$ rsync -a --progress rsync_example_dir ubuntu@146.235.208.193:~
sending incremental file list
rsync_example_dir/
rsync_example_dir/EXAMPLE_FILE
              0 100%    0.00kB/s    0:00:00 (xfr#1, to-chk=0/2)
```

--------------------------------

### Install Docker and NVIDIA Container Toolkit

Source: https://docs.lambda.ai/education/programming/virtual-environments-containers

Installs Docker and NVIDIA Container Toolkit on Ubuntu systems. This command should be run on instances where these are not pre-installed.

```bash
sudo apt -y update && sudo apt -y install docker.io nvidia-container-toolkit && \
sudo systemctl daemon-reload && \
sudo systemctl restart docker

```

--------------------------------

### Install prerequisites on Ubuntu

Source: https://docs.lambda.ai/education/scheduling-and-orchestration/orchestrating-workloads-with-dstack

Installs necessary system packages on Ubuntu for dstack and related tools. This includes python3, pip, git, curl, and jq.

```bash
sudo apt update && sudo apt install -y python3 python3-pip git curl jq
```

--------------------------------

### Install socat for Port Forwarding

Source: https://docs.lambda.ai/education/large-language-models/k8s-ollama-llama-3-2

Installs the socat utility, which is required for enabling port forwarding in subsequent steps of the tutorial.

```bash
sudo apt -y install socat

```

--------------------------------

### Install SkyPilot with Lambda and Kubernetes Support

Source: https://docs.lambda.ai/education/scheduling-and-orchestration/skypilot-deploy-kubernetes

Installs the SkyPilot package, including necessary extras for Lambda Cloud and Kubernetes integration.

```bash
pip3 install skypilot-nightly[lambda,kubernetes]

```

--------------------------------

### Install jq for Pretty-Printing JSON

Source: https://docs.lambda.ai/education/large-language-models/serving-llama-3-1-docker

Install the jq command-line JSON processor to format API responses. This is useful for making the output more human-readable.

```bash
sudo apt update && sudo apt install -y jq
```

--------------------------------

### Install K3s (Kubernetes)

Source: https://docs.lambda.ai/education/large-language-models/k8s-ollama-llama-3-2

Installs K3s, a lightweight Kubernetes distribution, with NVIDIA GPU support enabled. Ensure you are SSH'd into your instance before running.

```bash
curl -sfL https://get.k3s.io | K3S_KUBECONFIG_MODE=644 sh -s - --default-runtime=nvidia

```

--------------------------------

### Install Ollama Client

Source: https://docs.lambda.ai/education/large-language-models/k8s-ollama-llama-3-2

Download and install the Ollama client for Linux AMD64 systems. This involves downloading a tarball and extracting it to the system's binary directory.

```bash
curl -L https://ollama.com/download/ollama-linux-amd64.tgz -o ollama-linux-amd64.tgz
sudo tar -C /usr -xzf ollama-linux-amd64.tgz
```

--------------------------------

### Install Python Version with Full Support

Source: https://docs.lambda.ai/public-cloud/on-demand/getting-started

Installs a specific Python version alongside the preinstalled one, ensuring the 'venv' module is available for virtual environments. Append '-full' to the version number.

```bash
sudo apt -y update && sudo apt -y install pythonVERSION-full

```

```bash
sudo apt -y update && sudo apt -y install python3.13-full

```

--------------------------------

### Create and Navigate to Tutorial Directory

Source: https://docs.lambda.ai/education/scheduling-and-orchestration/skypilot-deploy-kubernetes

Sets up a dedicated directory for the SkyPilot tutorial and changes the current working directory into it.

```bash
mkdir ~/skypilot-tutorial && cd ~/skypilot-tutorial

```

--------------------------------

### Install ipmitool for Fan Control

Source: https://docs.lambda.ai/hardware/workstations/getting-started

Install the ipmitool utility to manage system hardware, including setting baseline fan speeds. This is required before using raw commands for fan control.

```bash
sudo apt -y update && sudo apt -y install ipmitool

```

--------------------------------

### Example PVC output

Source: https://docs.lambda.ai/public-cloud/1-click-clusters/managed-kubernetes/managed-kubernetes-legacy

This is an example of the output you might see when verifying PVCs using `kubectl get pvc`. It shows the PVC name, status, bound volume, capacity, access modes, and storage class.

```text
NAME                STATUS   VOLUME                                     CAPACITY   ACCESS MODES   STORAGECLASS    VOLUMEATTRIBUTESCLASS   AGE
huggingface-cache   Bound    pvc-8463f8d7-ca83-4dfd-8b21-a42edf09948b   400Gi      RWX            lambda-shared   <unset>                 45m

```

--------------------------------

### Run Llama 3.2 3B Model with Ollama

Source: https://docs.lambda.ai/education/large-language-models/k8s-ollama-llama-3-2

Execute this command in your terminal to start the Llama 3.2 3B model. Ensure Ollama is installed and configured.

```bash
ollama run llama3.2:3b-instruct-fp16
```

--------------------------------

### Download SkyPilot Kubernetes Deployment Files

Source: https://docs.lambda.ai/education/scheduling-and-orchestration/skypilot-deploy-kubernetes

Fetches the example configuration and launch script for deploying a Kubernetes cluster using SkyPilot.

```bash
curl -LO https://raw.githubusercontent.com/skypilot-org/skypilot/master/examples/k8s_cloud_deploy/cloud_k8s.yaml && \
    curl -LO https://raw.githubusercontent.com/skypilot-org/skypilot/master/examples/k8s_cloud_deploy/launch_k8s.sh

```

--------------------------------

### Navigate to SkyPilot Tutorial Directory

Source: https://docs.lambda.ai/education/scheduling-and-orchestration/skypilot-deploy-kubernetes

Change into the directory created for the SkyPilot tutorial.

```bash
cd ~/skypilot-tutorial
```

--------------------------------

### Launch Instance Request Sample

Source: https://docs.lambda.ai/api/cloud

Use this cURL command to launch a new instance. Ensure you replace '<YOUR-API-KEY>:' with your actual API key. The request body specifies instance configuration details.

```bash
curl --request POST --url 'https://cloud.lambda.ai/api/v1/instance-operations/launch' \
     --header 'accept: application/json' \
     --user '<YOUR-API-KEY>:' \
     --data '{ \
  "region_name": "string", \
  "instance_type_name": "gpu_8x_a100", \
  "ssh_key_names": [ \
    "my-public-key" \
  ], \
  "file_system_names": [ \
    "my-filesystem" \
  ], \
  "file_system_mounts": [ \
    { \
      "mount_point": "/data/custom-mount-point", \
      "file_system_id": "398578a2336b49079e74043f0bd2cfe8" \
    } \
  ], \
  "hostname": "headnode1", \
  "name": "My Instance", \
  "image": { \
    "id": "string" \
  }, \
  "user_data": "string", \
  "tags": [ \
    { \
      "key": "key1", \
      "value": "value1" \
    } \
  ], \
  "firewall_rulesets": [ \
    { \
      "id": "c4d291f47f9d436fa39f58493ce3b50d" \
    } \
  ] \
}'
```

--------------------------------

### Authenticate with curl using Basic Header

Source: https://docs.lambda.ai/api/cloud

Example of authenticating a GET request to list instances using `curl` with the Basic HTTP header. Replace `<YOUR-API-KEY>` with your actual API key.

```bash
curl --request GET --url 'https://cloud.lambda.ai/api/v1/instances' \
  --header 'accept: application/json' \
  --user '<YOUR-API-KEY>:'
```

--------------------------------

### Launch Instance API

Source: https://docs.lambda.ai/api/cloud

Launches a new instance with specified configurations. Requires authentication with an API key.

```APIDOC
## POST /api/v1/instance-operations/launch

### Description
Launches a new instance with specified configurations. Requires authentication with an API key.

### Method
POST

### Endpoint
https://cloud.lambda.ai/api/v1/instance-operations/launch

### Parameters
#### Request Body
- **region_name** (string) - Required - The AWS region where the instance will be launched.
- **instance_type_name** (string) - Required - The type of instance to launch (e.g., "gpu_8x_a100").
- **ssh_key_names** (array of string) - Required - A list of SSH key names to associate with the instance.
- **file_system_names** (array of string) - Optional - A list of file system names to attach to the instance.
- **file_system_mounts** (array of object) - Optional - A list of file system mount configurations.
  - **mount_point** (string) - Required - The mount point for the file system within the instance.
  - **file_system_id** (string) - Required - The ID of the file system to mount.
- **hostname** (string) - Optional - The hostname for the instance.
- **name** (string) - Required - The name of the instance.
- **image** (object) - Required - The image to use for the instance.
  - **id** (string) - Required - The ID of the image.
- **user_data** (string) - Optional - User data to be passed to the instance on launch.
- **tags** (array of object) - Optional - Tags to apply to the instance.
  - **key** (string) - Required - The tag key.
  - **value** (string) - Required - The tag value.
- **firewall_rulesets** (array of object) - Optional - Firewall rulesets to apply to the instance.
  - **id** (string) - Required - The ID of the firewall ruleset.

### Request Example
```json
{
  "region_name": "string",
  "instance_type_name": "gpu_8x_a100",
  "ssh_key_names": [
    "my-public-key"
  ],
  "file_system_names": [
    "my-filesystem"
  ],
  "file_system_mounts": [
    {
      "mount_point": "/data/custom-mount-point",
      "file_system_id": "398578a2336b49079e74043f0bd2cfe8"
    }
  ],
  "hostname": "headnode1",
  "name": "My Instance",
  "image": {
    "id": "string"
  },
  "user_data": "string",
  "tags": [
    {
      "key": "key1",
      "value": "value1"
    }
  ],
  "firewall_rulesets": [
    {
      "id": "c4d291f47f9d436fa39f58493ce3b50d"
    }
  ]
}
```

### Response
#### Success Response (200)
- **instance_ids** (array of string) - A list of the unique identifiers (IDs) of the launched instances.

#### Response Example
```json
{
  "data": {
    "instance_ids": [
      "0920582c7ff041399e34823a0be62549"
    ]
  }
}
```

#### Common Error Response Codes
- **400 Bad Request**: `instance-operations/launch/file-system-in-wrong-region`, `instance-operations/launch/insufficient-capacity`, `global/invalid-parameters`, `global/object-does-not-exist`, `global/quota-exceeded`
- **401 Unauthorized**: `global/invalid-api-key`
- **403 Forbidden**: `global/account-inactive`, `global/invalid-address`
- **404 Not Found**: `global/object-does-not-exist`
```

--------------------------------

### Install Mochi Dependencies

Source: https://docs.lambda.ai/education/fine-tune-mochi-gh200

Installs the Mochi repository and its dependencies, including specific versions of moviepy, pillow, and av. Ensure pip, setuptools, wheel, and packaging are up-to-date before installation. Requires sudo for apt package installation.

```bash
git clone https://github.com/genmoai/mochi.git
cd mochi-tune
pip install --upgrade pip setuptools wheel packaging
pip install -e . --no-build-isolation
pip install moviepy==1.0.3 pillow==9.5.0 av==13.1.0
sudo apt -y install bc
```

--------------------------------

### Clone and Prepare Repositories

Source: https://docs.lambda.ai/education/running-benchmark-gh200

Clone the DeepLearningExamples repository, checkout the 'lambda/benchmark' branch, and clone your forked deeplearning-benchmark repository. This sets up the necessary codebases for the benchmark.

```bash
git clone https://github.com/LambdaLabsML/DeepLearningExamples.git &&
cd DeepLearningExamples &&
git checkout lambda/benchmark &&
cd .. &&
git clone https://github.com/${GIT_USERNAME}/deeplearning-benchmark.git

```

--------------------------------

### Install Lambda Stack with cuDNN License Agreement

Source: https://docs.lambda.ai/education/linux-usage/lambda-stack-and-recovery-images

Installs the latest Lambda Stack on servers, explicitly agreeing to the NVIDIA cuDNN license. This is required for server installations.

```bash
wget -nv -O- https://lambda.ai/install-lambda-stack.sh | I_AGREE_TO_THE_CUDNN_LICENSE=1 sh - 

```

--------------------------------

### Apply Service Account Setup Manifest

Source: https://docs.lambda.ai/public-cloud/1-click-clusters/managed-kubernetes

Apply the service-account-setup.yaml manifest to create a Service Account, Secret, and ClusterRoleBinding. Ensure you replace placeholders like `<SERVICE-ACCOUNT-NAME>`, `<NAMESPACE>`, and `<ROLE>` before applying.

```bash
kubectl apply -f service-account-setup.yaml
```

--------------------------------

### PKCS8 Public Key Format Example

Source: https://docs.lambda.ai/public-cloud/on-demand/getting-started

An example of a PKCS8 public key format.

```text
-----BEGIN PUBLIC KEY-----
```

--------------------------------

### Install kubectl

Source: https://docs.lambda.ai/education/scheduling-and-orchestration/skypilot-deploy-kubernetes

Downloads and installs the latest stable version of kubectl for managing Kubernetes clusters.

```bash
curl -LO "https://dl.k8s.io/release/$(curl -L -s https://dl.k8s.io/release/stable.txt)/bin/linux/amd64/kubectl" && \
    sudo install -o root -g root -m 0755 kubectl /usr/local/bin/kubectl

```

--------------------------------

### Mount and Run Firmware Upgrade

Source: https://docs.lambda.ai/hardware/workstations/getting-started

Mounts the firmware ISO, extracts the upgrade utility, and executes it. This process requires root privileges and should be performed carefully.

```bash
mkdir /mnt/iso && mount -o loop Samsung_SSD_980_PRO_5B2QGXA7.iso /mnt/iso && \
mkdir fwupdate && cd fwupdate && \
gzip -dc /mnt/iso/initrd | cpio -idv --no-absolute-filenames && \
cd root/fumagician && ./fumagician
```

--------------------------------

### PEM Public Key Format Example

Source: https://docs.lambda.ai/public-cloud/on-demand/getting-started

An example of a PEM public key format, such as RSA.

```text
-----BEGIN RSA PUBLIC KEY-----
```

--------------------------------

### Start vLLM Cluster for Llama 3.1 Serving

Source: https://docs.lambda.ai/education/large-language-models/serving-llama-3-1-405b

Downloads a helper script and executes it to set up vLLM for multi-node inference and serving the Llama 3.1 405B model on the Lambda cluster.

```bash
curl -o "${SHARED_DIR}/run_cluster.sh" https://raw.githubusercontent.com/vllm-project/vllm/main/examples/run_cluster.sh
sudo bash "${SHARED_DIR}/run_cluster.sh" \
    vllm/vllm-openai \
    "${HEAD_IP}" \
    --head \
    "${HF_HOME}" \
    --privileged -e NCCL_IB_HCA=^mlx5_0
```

--------------------------------

### Download Miniconda Installer

Source: https://docs.lambda.ai/education/programming/virtual-environments-containers

Use this command to download the latest Miniconda3 installer script for Linux x86_64.

```bash
curl -O https://repo.anaconda.com/miniconda/Miniconda3-latest-Linux-x86_64.sh
```

--------------------------------

### List Filesystems Response Sample

Source: https://docs.lambda.ai/api/cloud

This is a sample JSON response when successfully listing filesystems. It includes details for each filesystem such as ID, name, mount point, creation date, creator information, usage status, region, and bytes used.

```json
{
  "data": [
    {
      "id": "398578a2336b49079e74043f0bd2cfe8",
      "name": "my-filesystem",
      "mount_point": "/lambda/nfs/my-filesystem",
      "created": "1970-01-01T00:00:00.000Z",
      "created_by": {
        "id": "3da5a70a57a7422ea8a7203f98b2198b",
        "email": "me@example.com",
        "status": "active"
      },
      "is_in_use": false,
      "region": {
        "name": "string",
        "description": "string"
      },
      "bytes_used": 0
    }
  ]
}
```

--------------------------------

### RFC4716 Public Key Format Example

Source: https://docs.lambda.ai/public-cloud/on-demand/getting-started

An example of an RFC4716 public key, often used by PuTTYgen.

```text
---- BEGIN SSH2 PUBLIC KEY ----
```

--------------------------------

### Add New User Account

Source: https://docs.lambda.ai/public-cloud/on-demand/managing-system-environment

Create a new user account on the system. This command prompts for user details and sets up a home directory.

```bash
sudo adduser <USERNAME>
```

--------------------------------

### Install smartmontools

Source: https://docs.lambda.ai/hardware/workstations/getting-started

Installs the smartmontools package required to check SSD firmware. Use this command on Debian-based systems.

```bash
sudo apt -y install smartmontools
```

--------------------------------

### Filesystem Mount Point Example

Source: https://docs.lambda.ai/public-cloud/on-demand

Illustrates the default mount point for a filesystem attached to an instance. The filesystem is also symlinked in the 'ubuntu' user's home directory.

```bash
/lambda/nfs/<FILESYSTEM_NAME>
```

--------------------------------

### Connect to Instance via SSH

Source: https://docs.lambda.ai/api/cloud

Establish an SSH connection to your launched instance using the generated private key. Replace `<INSTANCE-IP>` with the public IP address of your instance.

```bash
ssh -i key.pem <INSTANCE-IP>
```

--------------------------------

### Provider Error Example

Source: https://docs.lambda.ai/api/cloud

Example of an error originating from an upstream service, identified by the 'provider/' prefix in the error code.

```json
{
  "error": {
    "code": "provider/internal-unavailable",
    "message": "Provider unavailable",
    "suggestion": "Try again shortly"
  }
}
```

--------------------------------

### Update Terminal Environment

Source: https://docs.lambda.ai/public-cloud/on-demand/managing-system-environment

After installation, source your .bashrc file to apply changes made by the Miniconda installer to your terminal environment.

```bash
source ~/.bashrc
```

--------------------------------

### Install Nmap on Ubuntu

Source: https://docs.lambda.ai/public-cloud/on-demand/getting-started

Install the Nmap utility on your local Ubuntu machine. This tool is used to scan network ports.

```bash
sudo apt install -y nmap

```

--------------------------------

### Benchmark Nemotron 3 Nano with vLLM

Source: https://docs.lambda.ai/education/large-language-models/deploying-nemotron-3-nano

Run this command to benchmark Nemotron 3 Nano. It downloads a dataset, configures vLLM for serving, and specifies the model and backend. Ensure you have the dataset downloaded first.

```bash
wget https://huggingface.co/datasets/anon8231489123/ShareGPT_Vicuna_unfiltered/resolve/main/ShareGPT_V3_unfiltered_cleaned_split.json
vllm bench serve \
    --model nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16 \
    --served-model-name nemotron-3-nano  \
    --dataset-name sharegpt   \
    --dataset-path ./ShareGPT_V3_unfiltered_cleaned_split.json \
    --num-prompts 10 \
    --trust-remote-code \
    --backend openai-chat \
    --endpoint /v1/chat/completions

```

--------------------------------

### Install IPMI Tool

Source: https://docs.lambda.ai/hardware/servers/getting-started

Install the ipmitool package to manage IPMI functions on your server. This is a prerequisite for setting IPMI passwords.

```bash
sudo apt-get install ipmitool
```

--------------------------------

### List Available Software Modules

Source: https://docs.lambda.ai/public-cloud/1-click-clusters/managed-slurm

Use 'module avail' to see all software modules that can be loaded into your current environment.

```bash
module avail

```

--------------------------------

### Install FLUX.1 Model Dependencies

Source: https://docs.lambda.ai/education/generative-ai/flux-prompt-to-image

Install the necessary Python packages for serving the FLUX.1 model, including TensorRT support.

```bash
pip install -e '.[all,tensorrt]'

```

--------------------------------

### Delete example job

Source: https://docs.lambda.ai/public-cloud/1-click-clusters/managed-kubernetes

Run this command to immediately delete the example job. The job is typically configured to delete itself after completion.

```bash
kubectl delete -f https://docs.lambda.ai/assets/code/multiplication-eval-deepseek-r1-distilll-qwen-7b.yaml
```

--------------------------------

### Start vLLM Cluster on Head Node

Source: https://docs.lambda.ai/public-cloud/1-click-clusters/serving-llama-3_1-405b

Downloads a script to set up vLLM for multi-node inference and then runs it to start a Ray cluster for serving the Llama 3.1 405B model. It utilizes the InfiniBand fabric for optimal performance.

```bash
curl -o "${SHARED_DIR}/run_cluster.sh" https://raw.githubusercontent.com/vllm-project/vllm/main/examples/run_cluster.sh

sudo bash "${SHARED_DIR}/run_cluster.sh" \
     vllm/vllm-openai \
     "${HEAD_IP}" \
     --head "${HF_HOME}" \
     --privileged -e NCCL_IB_HCA=^mlx5_0
```

--------------------------------

### Example Nmap Scan Output

Source: https://docs.lambda.ai/public-cloud/on-demand/getting-started

This is an example of the output you might see after running an Nmap scan, indicating which ports are open.

```text
Starting Nmap 7.80 ( https://nmap.org ) at 2023-01-11 13:22 PST
Nmap scan report for 129.159.46.35
Host is up (0.041s latency).
Not shown: 999 filtered ports
PORT   STATE SERVICE
22/tcp open  ssh

Nmap done: 1 IP address (1 host up) scanned in 6.42 seconds

```

--------------------------------

### Open FortiClient VPN GUI

Source: https://docs.lambda.ai/private-cloud/managed-kubernetes/getting-started

Launches the FortiClient VPN graphical user interface from the terminal. Assumes FortiClient VPN is installed.

```bash
forticlient gui

```

--------------------------------

### Example Prompt for LLM Explanation

Source: https://docs.lambda.ai/education/large-language-models/kubeai-hermes-3

An example prompt to be used when interacting with large language models, requesting a detailed explanation of what they are.

```text
In 10,000 words or more, explain what large language models are.

```

--------------------------------

### Create and Activate Python Virtual Environment

Source: https://docs.lambda.ai/education/generative-ai/flux-prompt-to-image

Create and activate a Python virtual environment for the tutorial. This isolates project dependencies.

```bash
python3 -m venv .venv && source .venv/bin/activate

```