From 6876e3399c63c4e722de4381a3e24bdf8cab2c1e Mon Sep 17 00:00:00 2001 From: Mo King Date: Mon, 16 Mar 2026 15:07:55 -0400 Subject: [PATCH 1/7] add transformers lab --- docs.json | 3 +- integrations/transformer-lab.mdx | 241 +++++++++++++++++++++++++++++++ 2 files changed, 243 insertions(+), 1 deletion(-) create mode 100644 integrations/transformer-lab.mdx diff --git a/docs.json b/docs.json index 76fd1b0a..07fe2868 100644 --- a/docs.json +++ b/docs.json @@ -314,7 +314,8 @@ "pages":[ "integrations/dstack", "integrations/mods", - "integrations/skypilot" + "integrations/skypilot", + "integrations/transformer-lab" ] } ] diff --git a/integrations/transformer-lab.mdx b/integrations/transformer-lab.mdx new file mode 100644 index 00000000..5552748e --- /dev/null +++ b/integrations/transformer-lab.mdx @@ -0,0 +1,241 @@ +--- +title: "Run ML experiments on Runpod with Transformer Lab" +sidebarTitle: "Transformer Lab" +--- + +[Transformer Lab](https://lab.cloud/) is an open-source research environment for AI researchers to train, evaluate, and scale models from local hardware to cloud GPUs. It provides a unified interface for managing experiments, downloading models, and running training jobs across different compute providers. + +This guide shows you how to configure Transformer Lab to run ML workloads on Runpod GPUs using SkyPilot as the orchestration layer. + +## Requirements + +You'll need: + +* [A Runpod account with an API key](/get-started/api-keys). +* macOS, Linux, or Windows with WSL2. +* Python 3.8 or higher. +* Git and curl installed. + + + +**Windows users** + +Transformer Lab requires [WSL2 (Windows Subsystem for Linux)](https://docs.microsoft.com/en-us/windows/wsl/install). Install WSL2 first, then follow the Linux instructions within your WSL2 environment. + + + +## Install Transformer Lab + + + + Open a terminal and run: + + ```bash + curl -LsSf https://lab.cloud/install.sh | bash + ``` + + This installs Transformer Lab to `~/.transformerlab` and sets up a conda environment with all dependencies. + + + + To enable cloud compute providers like Runpod, run the multiuser setup: + + ```bash + ~/.transformerlab/src/install.sh multiuser_setup + ``` + + This installs SkyPilot and enables the Team Settings features needed for cloud provider configuration. + + + + Start the Transformer Lab server: + + ```bash + cd ~/.transformerlab/src + ./run.sh + ``` + + Open your browser to `http://localhost:8338`. + + + + Use the default credentials: + + * **Email**: `admin@example.com` + * **Password**: `admin123` + + + Change these credentials after your first login for security. + + + + +## Configure Runpod as a compute provider + + + + In the Runpod console, go to [Settings](https://www.console.runpod.io/user/settings) and create an API key with **All** permissions or **Restricted** permissions that include Pod access. + + Copy the API key. Runpod doesn't store it, so save it securely. + + + + In Transformer Lab, click your profile icon in the top right corner and select **Team Settings**. + + + + Navigate to **Compute Providers** and click **Add Provider**. + + In the modal that opens: + + 1. Enter a name for your provider (e.g., "Runpod GPUs"). + 2. Select **Runpod (beta)** as the provider type. + 3. In the configuration JSON field, add your [Runpod API key](/get-started/api-keys): + + ```json + { + "api_key": "YOUR_RUNPOD_API_KEY", + "api_base_url": "https://rest.runpod.io/v1" + } + ``` + + Leave the base URL as is. + + You can also set optional defaults: + + ```json + { + "api_key": "YOUR_RUNPOD_API_KEY", + "default_gpu_type": "NVIDIA GeForce RTX 4090", + "default_region": "US" + } + ``` + + Click **Add Compute Provider** to save the provider. + + + +## Run a task on Runpod + +Transformer Lab uses SkyPilot to orchestrate cloud workloads. You define tasks in YAML files that specify the resources, setup commands, and run commands. + +### Create a task file + +Create a file named `hello_runpod.yaml`: + +```yaml +resources: + cloud: runpod + accelerators: A40:1 + +setup: | + echo "Setting up environment..." + pip install torch + +run: | + echo "Hello from Runpod!" + python << 'EOF' + import torch + print(f"CUDA available: {torch.cuda.is_available()}") + if torch.cuda.is_available(): + print(f"GPU: {torch.cuda.get_device_name(0)}") + else: + print("GPU: None") + EOF +``` + +### Launch the task + +Launch your task on Runpod: + +```bash +sky launch -c my-cluster hello_runpod.yaml +``` + +SkyPilot provisions a Pod on Runpod, runs your task, and streams the output to your terminal. + +### View and manage clusters + +List running clusters: + +```bash +sky status +``` + +SSH into a cluster: + +```bash +ssh my-cluster +``` + +Stop a cluster when you're done: + +```bash +sky down my-cluster +``` + +## Run a training job + +Here's an example task file for fine-tuning a model on Runpod: + +```yaml +resources: + cloud: runpod + accelerators: RTX4090:1 + disk_size: 100 + +setup: | + pip install torch transformers datasets accelerate peft + +run: | + python << 'EOF' + import torch + from transformers import AutoModelForCausalLM, AutoTokenizer, TrainingArguments + from peft import LoraConfig, get_peft_model + + print(f"GPU: {torch.cuda.get_device_name(0)}") + print(f"Memory: {torch.cuda.get_device_properties(0).total_memory / 1e9:.1f} GB") + + # Load model + model_name = "HuggingFaceTB/SmolLM-135M-Instruct" + model = AutoModelForCausalLM.from_pretrained(model_name) + tokenizer = AutoTokenizer.from_pretrained(model_name) + + # Configure LoRA + lora_config = LoraConfig( + r=8, + lora_alpha=32, + target_modules=["q_proj", "v_proj"], + lora_dropout=0.05, + ) + model = get_peft_model(model, lora_config) + + print("Model ready for training!") + print(f"Trainable parameters: {model.print_trainable_parameters()}") + EOF +``` + +Launch the training job: + +```bash +sky launch -c training-cluster finetune.yaml +``` + +## Available GPU types + +Runpod offers a variety of GPUs. Use the `accelerators` field to specify the GPU type: + +| Accelerator | Description | +|-------------|-------------| +| `RTX4090:1` | NVIDIA GeForce RTX 4090 (24GB) | +| `A40:1` | NVIDIA A40 (48GB) | +| `A100:1` | NVIDIA A100 (40GB or 80GB) | +| `A100-80GB:1` | NVIDIA A100 80GB | +| `H100:1` | NVIDIA H100 (80GB) | +| `L40S:1` | NVIDIA L40S (48GB) | + +For multiple GPUs, change the count: `A100:4` for 4x A100 GPUs. + +## Clean up + +To avoid unexpected charges, stop your Pods using the [Runpod console](https://www.console.runpod.io/pods) when you're done. From 368226fe821672da8d5230e6828d07b1473b6229 Mon Sep 17 00:00:00 2001 From: Mo King Date: Mon, 16 Mar 2026 16:37:05 -0400 Subject: [PATCH 2/7] Transformer Lab vs. --- integrations/transformer-lab.mdx | 125 +++++++++++++------------------ 1 file changed, 54 insertions(+), 71 deletions(-) diff --git a/integrations/transformer-lab.mdx b/integrations/transformer-lab.mdx index 5552748e..8612c2e0 100644 --- a/integrations/transformer-lab.mdx +++ b/integrations/transformer-lab.mdx @@ -5,7 +5,7 @@ sidebarTitle: "Transformer Lab" [Transformer Lab](https://lab.cloud/) is an open-source research environment for AI researchers to train, evaluate, and scale models from local hardware to cloud GPUs. It provides a unified interface for managing experiments, downloading models, and running training jobs across different compute providers. -This guide shows you how to configure Transformer Lab to run ML workloads on Runpod GPUs using SkyPilot as the orchestration layer. +This guide shows you how to configure Transformer Lab to run ML workloads on Runpod GPUs. ## Requirements @@ -117,66 +117,64 @@ Transformer Lab requires [WSL2 (Windows Subsystem for Linux)](https://docs.micro ## Run a task on Runpod -Transformer Lab uses SkyPilot to orchestrate cloud workloads. You define tasks in YAML files that specify the resources, setup commands, and run commands. +Transformer Lab uses task files to define cloud workloads. Tasks specify the resources, setup commands, and run commands for your job. -### Create a task file +### Create a task -Create a file named `hello_runpod.yaml`: + + + In the Transformer Lab sidebar, click **Tasks** to open the task management interface. + -```yaml -resources: - cloud: runpod - accelerators: A40:1 + + Click **New** to add a new task. Select **Start with a blank task template**, then click **Submit**. + -setup: | - echo "Setting up environment..." - pip install torch + + In the task editor, paste the following YAML configuration: -run: | - echo "Hello from Runpod!" - python << 'EOF' - import torch - print(f"CUDA available: {torch.cuda.is_available()}") - if torch.cuda.is_available(): - print(f"GPU: {torch.cuda.get_device_name(0)}") - else: - print("GPU: None") - EOF -``` + ```yaml + resources: + cloud: runpod + accelerators: A40:1 -### Launch the task + setup: | + echo "Setting up environment..." + pip install torch -Launch your task on Runpod: + run: | + echo "Hello from Runpod!" + python -c "import torch; print(f'CUDA available: {torch.cuda.is_available()}'); print(f'GPU: {torch.cuda.get_device_name(0)}' if torch.cuda.is_available() else 'GPU: None')" + ``` -```bash -sky launch -c my-cluster hello_runpod.yaml -``` + This configuration requests a single NVIDIA A40 GPU on Runpod, installs PyTorch, and runs a simple script to verify GPU access. + -SkyPilot provisions a Pod on Runpod, runs your task, and streams the output to your terminal. + + Click **Queue** to submit the task you just created. Select your Runpod compute provider and click **Submit** to start the job. Transformer Lab provisions a Pod on Runpod, runs your task, and displays the output in the task logs. + + -### View and manage clusters +### Monitor task progress -List running clusters: +Once queued, your task appears in the Tasks list with its current status. Click **Output** to view the task logs. -```bash -sky status -``` +The output modal has two tabs: -SSH into a cluster: +- **Lab SDK Output**: Shows output from scripts that use the `transformerlab` Python package. +- **Machine Logs**: Shows raw stdout/stderr from the Pod. Use this tab to see output from standard `print()` statements. -```bash -ssh my-cluster -``` +For the examples in this guide, check the **Machine Logs** tab to see your task output. -Stop a cluster when you're done: +### Stop a running task -```bash -sky down my-cluster -``` +To stop a task before it completes, click the stop button (square icon). This terminates the Runpod Pod and releases the resources. + +You can also verify that no Pods are running by checking the [Runpod console](https://www.console.runpod.io/pods). ## Run a training job -Here's an example task file for fine-tuning a model on Runpod: +Here's an example task configuration for fine-tuning a model on Runpod. Create a new task in the Tasks menu and use this YAML: ```yaml resources: @@ -185,41 +183,24 @@ resources: disk_size: 100 setup: | - pip install torch transformers datasets accelerate peft + pip install torch transformers peft run: | - python << 'EOF' + python -c " import torch - from transformers import AutoModelForCausalLM, AutoTokenizer, TrainingArguments + from transformers import AutoModelForCausalLM, AutoTokenizer from peft import LoraConfig, get_peft_model - - print(f"GPU: {torch.cuda.get_device_name(0)}") - print(f"Memory: {torch.cuda.get_device_properties(0).total_memory / 1e9:.1f} GB") - - # Load model - model_name = "HuggingFaceTB/SmolLM-135M-Instruct" - model = AutoModelForCausalLM.from_pretrained(model_name) - tokenizer = AutoTokenizer.from_pretrained(model_name) - - # Configure LoRA - lora_config = LoraConfig( - r=8, - lora_alpha=32, - target_modules=["q_proj", "v_proj"], - lora_dropout=0.05, - ) + print(f'GPU: {torch.cuda.get_device_name(0)}') + print(f'Memory: {torch.cuda.get_device_properties(0).total_memory / 1e9:.1f} GB') + model = AutoModelForCausalLM.from_pretrained('HuggingFaceTB/SmolLM-135M-Instruct') + lora_config = LoraConfig(r=8, lora_alpha=32, target_modules=['q_proj', 'v_proj'], lora_dropout=0.05) model = get_peft_model(model, lora_config) - - print("Model ready for training!") - print(f"Trainable parameters: {model.print_trainable_parameters()}") - EOF + print('Model ready for training!') + model.print_trainable_parameters() + " ``` -Launch the training job: - -```bash -sky launch -c training-cluster finetune.yaml -``` +Select your Runpod compute provider and click **Queue** to start the training job. ## Available GPU types @@ -238,4 +219,6 @@ For multiple GPUs, change the count: `A100:4` for 4x A100 GPUs. ## Clean up -To avoid unexpected charges, stop your Pods using the [Runpod console](https://www.console.runpod.io/pods) when you're done. +When your tasks complete, Transformer Lab automatically releases the Runpod resources. To manually stop a running task, select it from the Tasks list and click **Stop**. + +You can also verify that no Pods are running by checking the [Runpod console](https://www.console.runpod.io/pods). From 3747286ad1e01bcd3ee2737bdf25e5152404c6bc Mon Sep 17 00:00:00 2001 From: Mo King Date: Tue, 17 Mar 2026 10:01:15 -0400 Subject: [PATCH 3/7] Update transformer lab --- docs.json | 4 +- integrations/transformer-lab.mdx | 87 +++++++++++--------------------- 2 files changed, 32 insertions(+), 59 deletions(-) diff --git a/docs.json b/docs.json index df3cc1f1..1305a972 100644 --- a/docs.json +++ b/docs.json @@ -249,10 +249,10 @@ { "group": "Guides", "pages":[ + "integrations/transformer-lab", "integrations/dstack", "integrations/mods", - "integrations/skypilot", - "integrations/transformer-lab" + "integrations/skypilot" ] } ] diff --git a/integrations/transformer-lab.mdx b/integrations/transformer-lab.mdx index 8612c2e0..0078cdd5 100644 --- a/integrations/transformer-lab.mdx +++ b/integrations/transformer-lab.mdx @@ -44,7 +44,7 @@ Transformer Lab requires [WSL2 (Windows Subsystem for Linux)](https://docs.micro ~/.transformerlab/src/install.sh multiuser_setup ``` - This installs SkyPilot and enables the Team Settings features needed for cloud provider configuration. + This enables the Team Settings features needed for cloud provider configuration. @@ -70,6 +70,16 @@ Transformer Lab requires [WSL2 (Windows Subsystem for Linux)](https://docs.micro +## Configure shared storage + +For remote task execution, Transformer Lab requires shared storage so your local instance can communicate with remote Pods. Configure one of the following: + +- **Amazon S3**: Create an S3 bucket and configure credentials +- **Google Cloud Storage**: Create a GCS bucket and configure service account +- **Azure Blob Storage**: Create a storage container and configure credentials + +Refer to the [Transformer Lab documentation](https://lab.cloud/for-teams/) for detailed shared storage setup instructions. + ## Configure Runpod as a compute provider @@ -88,7 +98,7 @@ Transformer Lab requires [WSL2 (Windows Subsystem for Linux)](https://docs.micro In the modal that opens: - 1. Enter a name for your provider (e.g., "Runpod GPUs"). + 1. Enter a name for your provider (e.g., "runpod-provider"). Remember this name—you'll use it in your task.yaml files. 2. Select **Runpod (beta)** as the provider type. 3. In the configuration JSON field, add your [Runpod API key](/get-started/api-keys): @@ -99,18 +109,8 @@ Transformer Lab requires [WSL2 (Windows Subsystem for Linux)](https://docs.micro } ``` - Leave the base URL as is. + Leave the base URL as is. - You can also set optional defaults: - - ```json - { - "api_key": "YOUR_RUNPOD_API_KEY", - "default_gpu_type": "NVIDIA GeForce RTX 4090", - "default_region": "US" - } - ``` - Click **Add Compute Provider** to save the provider. @@ -134,19 +134,22 @@ Transformer Lab uses task files to define cloud workloads. Tasks specify the res In the task editor, paste the following YAML configuration: ```yaml + name: hello-runpod resources: - cloud: runpod - accelerators: A40:1 - + compute_provider: runpod-provider + cpus: 4 + memory: 16 + accelerators: "A40:1" setup: | echo "Setting up environment..." pip install torch - run: | echo "Hello from Runpod!" python -c "import torch; print(f'CUDA available: {torch.cuda.is_available()}'); print(f'GPU: {torch.cuda.get_device_name(0)}' if torch.cuda.is_available() else 'GPU: None')" ``` + Replace `runpod-provider` with the name you gave your Runpod provider in Team Settings. + This configuration requests a single NVIDIA A40 GPU on Runpod, installs PyTorch, and runs a simple script to verify GPU access. @@ -172,50 +175,20 @@ To stop a task before it completes, click the stop button (square icon). This te You can also verify that no Pods are running by checking the [Runpod console](https://www.console.runpod.io/pods). -## Run a training job - -Here's an example task configuration for fine-tuning a model on Runpod. Create a new task in the Tasks menu and use this YAML: - -```yaml -resources: - cloud: runpod - accelerators: RTX4090:1 - disk_size: 100 - -setup: | - pip install torch transformers peft - -run: | - python -c " - import torch - from transformers import AutoModelForCausalLM, AutoTokenizer - from peft import LoraConfig, get_peft_model - print(f'GPU: {torch.cuda.get_device_name(0)}') - print(f'Memory: {torch.cuda.get_device_properties(0).total_memory / 1e9:.1f} GB') - model = AutoModelForCausalLM.from_pretrained('HuggingFaceTB/SmolLM-135M-Instruct') - lora_config = LoraConfig(r=8, lora_alpha=32, target_modules=['q_proj', 'v_proj'], lora_dropout=0.05) - model = get_peft_model(model, lora_config) - print('Model ready for training!') - model.print_trainable_parameters() - " -``` - -Select your Runpod compute provider and click **Queue** to start the training job. - -## Available GPU types +## Specify GPU types -Runpod offers a variety of GPUs. Use the `accelerators` field to specify the GPU type: +Use the `accelerators` field to specify the GPU type: | Accelerator | Description | |-------------|-------------| -| `RTX4090:1` | NVIDIA GeForce RTX 4090 (24GB) | -| `A40:1` | NVIDIA A40 (48GB) | -| `A100:1` | NVIDIA A100 (40GB or 80GB) | -| `A100-80GB:1` | NVIDIA A100 80GB | -| `H100:1` | NVIDIA H100 (80GB) | -| `L40S:1` | NVIDIA L40S (48GB) | - -For multiple GPUs, change the count: `A100:4` for 4x A100 GPUs. +| `"RTX4090:1"` | NVIDIA GeForce RTX 4090 (24GB) | +| `"A40:1"` | NVIDIA A40 (48GB) | +| `"A100:1"` | NVIDIA A100 (40GB or 80GB) | +| `"A100-80GB:1"` | NVIDIA A100 80GB | +| `"H100:1"` | NVIDIA H100 (80GB) | +| `"L40S:1"` | NVIDIA L40S (48GB) | + +For multiple GPUs, change the count: `"A100:4"` for 4x A100 GPUs. ## Clean up From c4ae4fc7e3c84cc4035cb374a7f07ae4bdbbe20c Mon Sep 17 00:00:00 2001 From: Mo King Date: Tue, 17 Mar 2026 10:05:25 -0400 Subject: [PATCH 4/7] Add links to TL docs --- integrations/transformer-lab.mdx | 2 ++ 1 file changed, 2 insertions(+) diff --git a/integrations/transformer-lab.mdx b/integrations/transformer-lab.mdx index 0078cdd5..e8d26d43 100644 --- a/integrations/transformer-lab.mdx +++ b/integrations/transformer-lab.mdx @@ -119,6 +119,8 @@ Refer to the [Transformer Lab documentation](https://lab.cloud/for-teams/) for d Transformer Lab uses task files to define cloud workloads. Tasks specify the resources, setup commands, and run commands for your job. +For detailed information on task configuration, see the [Task YAML Structure](https://lab.cloud/for-teams/running-a-task/task-yaml-structure) documentation. You can also browse the [Task Gallery](https://lab.cloud/for-teams/running-a-task/quick-start#4-import-a-task-from-tasks-gallery) for pre-built templates. + ### Create a task From 114df12933dea85a4989ae0d2cb2b85bb9e718e0 Mon Sep 17 00:00:00 2001 From: Mo King Date: Tue, 17 Mar 2026 10:12:24 -0400 Subject: [PATCH 5/7] Add description metadata to integration guides --- integrations/dstack.mdx | 1 + integrations/mods.mdx | 1 + integrations/skypilot.mdx | 1 + integrations/transformer-lab.mdx | 1 + 4 files changed, 4 insertions(+) diff --git a/integrations/dstack.mdx b/integrations/dstack.mdx index 8e1c5558..c917e1ee 100644 --- a/integrations/dstack.mdx +++ b/integrations/dstack.mdx @@ -1,6 +1,7 @@ --- title: "Manage Pods with dstack on Runpod" sidebarTitle: "dstack" +description: "Use dstack to automate Pod orchestration for AI and ML workloads on Runpod." --- [dstack](https://dstack.ai/) is an open-source tool that automates Pod orchestration for AI and ML workloads. It lets you define your application and resource requirements in YAML files, then handles provisioning and managing cloud resources on Runpod so you can focus on your application instead of infrastructure. diff --git a/integrations/mods.mdx b/integrations/mods.mdx index 1000b1c5..92d04347 100644 --- a/integrations/mods.mdx +++ b/integrations/mods.mdx @@ -1,6 +1,7 @@ --- title: "Running Runpod on Mods" sidebarTitle: "Mods" +description: "Use Mods to interact with language models hosted on Runpod from the command line." --- [Mods](https://github.com/charmbracelet/mods) is a command-line tool for interacting with language models. It integrates with Unix pipelines, letting you send command output directly to LLMs from your terminal. diff --git a/integrations/skypilot.mdx b/integrations/skypilot.mdx index 853f3304..eb19af9e 100644 --- a/integrations/skypilot.mdx +++ b/integrations/skypilot.mdx @@ -1,6 +1,7 @@ --- title: "Running Runpod on SkyPilot" sidebarTitle: "SkyPilot" +description: "Use SkyPilot to run LLMs, AI, and batch jobs on Runpod Pods and Serverless endpoints." --- [SkyPilot](https://skypilot.readthedocs.io/en/latest/) is a framework for running LLMs, AI, and batch jobs on any cloud. diff --git a/integrations/transformer-lab.mdx b/integrations/transformer-lab.mdx index e8d26d43..088b5a3f 100644 --- a/integrations/transformer-lab.mdx +++ b/integrations/transformer-lab.mdx @@ -1,6 +1,7 @@ --- title: "Run ML experiments on Runpod with Transformer Lab" sidebarTitle: "Transformer Lab" +description: "Configure Transformer Lab to run ML training and inference workloads on Runpod GPUs." --- [Transformer Lab](https://lab.cloud/) is an open-source research environment for AI researchers to train, evaluate, and scale models from local hardware to cloud GPUs. It provides a unified interface for managing experiments, downloading models, and running training jobs across different compute providers. From 74b9289e7d972729737a62fcd681697bf1259956 Mon Sep 17 00:00:00 2001 From: Mo King Date: Thu, 19 Mar 2026 10:29:24 -0400 Subject: [PATCH 6/7] Update TL integration doc with feedback from TL team --- integrations/transformer-lab.mdx | 22 ++++++---------------- 1 file changed, 6 insertions(+), 16 deletions(-) diff --git a/integrations/transformer-lab.mdx b/integrations/transformer-lab.mdx index 088b5a3f..a8862dcc 100644 --- a/integrations/transformer-lab.mdx +++ b/integrations/transformer-lab.mdx @@ -4,7 +4,7 @@ sidebarTitle: "Transformer Lab" description: "Configure Transformer Lab to run ML training and inference workloads on Runpod GPUs." --- -[Transformer Lab](https://lab.cloud/) is an open-source research environment for AI researchers to train, evaluate, and scale models from local hardware to cloud GPUs. It provides a unified interface for managing experiments, downloading models, and running training jobs across different compute providers. +[Transformer Lab](https://lab.cloud/) is an open-source research environment for AI researchers to train, fine-tune and evaluate models. It allows you to easily scale training from local hardware to cloud GPUs. It provides a unified interface to all your compute resources and simplifies experiment/checkpoint tracking, job scheduling, auto-recovery, centralized artifact storage and more. This guide shows you how to configure Transformer Lab to run ML workloads on Runpod GPUs. @@ -17,13 +17,13 @@ You'll need: * Python 3.8 or higher. * Git and curl installed. - + **Windows users** Transformer Lab requires [WSL2 (Windows Subsystem for Linux)](https://docs.microsoft.com/en-us/windows/wsl/install). Install WSL2 first, then follow the Linux instructions within your WSL2 environment. - + ## Install Transformer Lab @@ -32,20 +32,10 @@ Transformer Lab requires [WSL2 (Windows Subsystem for Linux)](https://docs.micro Open a terminal and run: ```bash - curl -LsSf https://lab.cloud/install.sh | bash + curl -fsSL https://lab.cloud/install.sh | bash -s -- multiuser_setup ``` - This installs Transformer Lab to `~/.transformerlab` and sets up a conda environment with all dependencies. - - - - To enable cloud compute providers like Runpod, run the multiuser setup: - - ```bash - ~/.transformerlab/src/install.sh multiuser_setup - ``` - - This enables the Team Settings features needed for cloud provider configuration. + This installs Transformer Lab to `~/.transformerlab`, sets up a conda environment with all dependencies, and enables the Team Settings features needed for cloud provider configuration. @@ -100,7 +90,7 @@ Refer to the [Transformer Lab documentation](https://lab.cloud/for-teams/) for d In the modal that opens: 1. Enter a name for your provider (e.g., "runpod-provider"). Remember this name—you'll use it in your task.yaml files. - 2. Select **Runpod (beta)** as the provider type. + 2. Select **Runpod** as the provider type. 3. In the configuration JSON field, add your [Runpod API key](/get-started/api-keys): ```json From f3904c122c400cd1146db9a1a1517cc89af76e9b Mon Sep 17 00:00:00 2001 From: Mo King Date: Thu, 19 Mar 2026 10:33:46 -0400 Subject: [PATCH 7/7] Link to shared storage documentation --- integrations/transformer-lab.mdx | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/integrations/transformer-lab.mdx b/integrations/transformer-lab.mdx index a8862dcc..b91e1a8d 100644 --- a/integrations/transformer-lab.mdx +++ b/integrations/transformer-lab.mdx @@ -65,11 +65,11 @@ Transformer Lab requires [WSL2 (Windows Subsystem for Linux)](https://docs.micro For remote task execution, Transformer Lab requires shared storage so your local instance can communicate with remote Pods. Configure one of the following: -- **Amazon S3**: Create an S3 bucket and configure credentials -- **Google Cloud Storage**: Create a GCS bucket and configure service account -- **Azure Blob Storage**: Create a storage container and configure credentials +- **Amazon S3**: Create an S3 bucket and configure credentials. +- **Google Cloud Storage**: Create a GCS bucket and configure service account. +- **Azure Blob Storage**: Create a storage container and configure credentials. -Refer to the [Transformer Lab documentation](https://lab.cloud/for-teams/) for detailed shared storage setup instructions. +Refer to the [Transformer Lab documentation](https://lab.cloud/for-teams/advanced-install/cloud-storage/) for detailed shared storage setup instructions. ## Configure Runpod as a compute provider