Contact us if you'd like to request a custom solution or instance type.

Amazon Web Services

Microsoft Azure

Google Cloud Platform

CPU GPU INF2

Nvidia RTX PRO 6000 Blackwell

2x GPUs · 192 GB 46x vCPUs · 512 GB

$5.5 / h

available Hardware should be compatible with the selected model.

Private Public Authenticated

Only you can access your endpoint, using a Hugging Face Token generated from your personal account.

Automatic Scale-to-Zero

Endpoints scaled to 0 replicas are not billed. They may take some time to scale back up once they start receiving requests again.

Number of replicas

Automatically scale the number of replicas within Min and Max based on compute usage. Min is always 0 if Scale-To-Zero is active.

Min Max

More options

Autoscaling Strategy

Control what type of trigger will cause your Endpoint to scale up.

Hardware Usage Pending Requests

Hardware Utilization Threshold (%)

A scale up event will be triggered if the average hardware utilisation (%) exceeds this threshold for more than 20 seconds.

Inference Engine

Engine Container URL optional

Max Batched Tokens

Maximum number of tokens to be processed in a single iteration. If left unspecified by the user, vLLM will be set it based on the usage context.

optional

Max Sequences

Maximum number of sequences to be processed in a single iteration. If left unspecified by the user, vLLM will be set it based on the usage context.

optional

Tensor Parallel Size

Number of tensor parallel groups for distributed inference

optional

Data Parallel Size

Number of data parallel groups for distributed inference

optional

KV Cache Data Type

KV cache data type for memory optimization

optional

Server Arguments

Arguments passed to the engine server entrypoint.

optional

Server Command

Command executed as engine server entrypoint.

optional

Default Env

Environment variables that will be provided to your container during deployment.

Secret Env

Same as Default, but people with access to this endpoint will not be able to read these values after creation.

VPC Config

Check to activate and configure AWS PrivateLink

Use AWS PrivateLink

Commit Revision

Specify a revision commit hash for the Hugging Face repository

optional

Task

Select a supported task with the right inputs and outputs for your model pipeline, or define a custom task.

Download Pattern

Specify what file type(s) should be downloaded from the repository.

Inkling-Small-NVFP4