Amazon SageMaker Studio now lets users launch a development environment with a single click from a Hugging Face model page.
Previously, finding a model on Hugging Face meant navigating to the AWS Management Console to create a domain, configure Identity and Access Management (IAM) permissions, and request graphics processing unit (GPU) quotas. This process slowed the move from discovery to experimentation. The new integration removes that friction.
“At Arcee, we build open models so developers and enterprises can actually own what they run: inspect the weights, post-train on their own data, and deploy on their own terms. This integration takes that promise the last mile. Going from an open model on Hugging Face straight into SageMaker Studio in a single click, then fine-tuning or deploying it inside your own AWS environment with nothing to wire up, is the kind of experience open models have been missing. Open weights you own, running in the cloud you control. That is exactly the combination our customers have been asking for.”
Mark McQuade, Founder and CEO of Arcee AI, said this.
The new one-click Studio landing experience appears when a user selects Customize on SageMaker AI or Deploy on SageMaker AI on a supported Hugging Face model page. This action sends the user directly to the console. SageMaker AI then automatically provisions a new domain with pre-configured permissions in seconds and maintains the model context.
What’s new
The launch adds three features that shorten the path from a Hugging Face model to a working SageMaker Studio workflow.
Deep links from Hugging Face into SageMaker Studio
Browsing models on Hugging Face now shows action buttons next to supported models that map directly to SageMaker Studio workflows:
- Customize on SageMaker AI opens the Model Customization page in Studio with the selected model pre-loaded, ready to fine-tune.
- Deploy on SageMaker AI opens the Deployment page in Studio with the model pre-configured for endpoint deployment.
Each entry point preserves the context, meaning users do not need to search for the model again once inside Studio.
Pre-configured permissions
New Studio environments created through this flow come with permissions already configured for the full range of SageMaker AI capabilities, including model customization, training jobs, notebook experimentation, and endpoint deployment. A new managed policy, AmazonSageMakerModelCustomizationCoreAccess, is created and attached for the user. It provides permissions for serverless model customization jobs using supervised fine-tuning (SFT), direct preference optimization (DPO), reinforcement learning with verifiable rewards (RLVR), and reinforcement learning from AI feedback (RLAIF), with supported deployment to SageMaker AI or Amazon Bedrock endpoints. This removes the need to manually create and configure AWS Identity and Access Management (IAM) roles and policies before starting to experiment. For existing Studio environments, actionable messages with direct links to documentation guide users through adding these permissions.
GPU quota visibility
When selecting instance types for deployment or training, the Studio UI now shows quota availability directly in the instance selection list. Users can immediately see which GPU instance types (G5, G6) are available under their account’s current limits. There is no need to navigate separately to Service Quotas. If a limit increase is still required, the system redirects the user directly to the Service Quotas page for the respective instance type.
Walkthrough: Deep-linking from Hugging Face to SageMaker Studio
This section walks through customizing or deploying a model starting from Hugging Face.
Step 1: Discover and select
On the Hugging Face model page, select Customize on SageMaker AI for a supported model.
Step 2: Sign in
Users are prompted to sign in to AWS using their existing credentials. If an active console session already exists, this step is skipped automatically. For more information, see Sign in to the AWS Management Console.
Step 3: Land in Studio
Users arrive directly on the Model Customization page inside SageMaker Studio with their model pre-selected. Next, they configure their fine-tuning parameters such as training data, hyperparameters, and instance type, then submit the customization job.
Alternatively, selecting Deploy on SageMaker AI opens the endpoint deployment page in Studio with the model pre-configured. Users select their instance type (quota visibility included), review the settings, and deploy.
Step 4: Test your endpoint
After deploying the endpoint, users test inference directly from Studio’s endpoint testing interface.
Getting started
Users can try this experience today:
- Browse models on Hugging Face.
- Look for the Customize on SageMaker AI or Deploy on SageMaker AI buttons on supported models.
- Select and follow the streamlined sign-in flow.
- Start building in a fully configured SageMaker Studio environment.
What it means
The change reduces the steps between finding a model and using it. Developers can stay in their current flow without context switching or manual environment setup. There is no need to troubleshoot permissions.
To get started, visit the Amazon SageMaker Studio page or explore models on Hugging Face and choose Deploy or Customize on SageMaker AI.
