---
title: "Get a container deployment template"
method: GET
path: "/v1/container-deployment-templates/{template_id}"
tags: ["Serverless Containers"]
---

# Get a container deployment template

`GET /v1/container-deployment-templates/{template_id}`

## Request

- Base URL: `https://api.verda.com`
- URL: `https://api.verda.com/v1/container-deployment-templates/{template_id}`
- Auth: HTTP bearer

## Path parameters

- `template_id` string, required

## Response `200`

The template with its variants and features

- ContainerDeploymentTemplateDetailPublicApiDto
  - `id` string, required: Stable, URL-safe template identifier used in template paths
  - `name` string, required: Human-readable name of the model
  - `description` string, required: Description of the model
  - `provider` string, required: Provider of the model
  - `engine` string, required: Serving engine used by the template
  - `model_repository` string, required: Hugging Face repository the default variant loads
  - `parameters` string, required: Parameter count of the model (display string)
  - `context_length` number, required: Maximum context length in tokens
  - `input_modalities` string[], required: Accepted input modalities
  - `output_modalities` string[], required: Produced output modalities
  - `tasks` string[], required: Coarse task labels
  - `requires_hugging_face_token` boolean, required: Whether a Hugging Face token is required to deploy the model (gated checkpoint)
  - `min_vram_gb` number, required: Minimum total GPU memory in GB for the default variant
  - `recommended_compute` ComputeResource, required
    - `name` string, required: Name of the compute resource
    - `size` number, required: Number of compute units (e.g. 4 GPUs). Default is 1
  - `revision` object, nullable, required: Pinned Hugging Face revision of the default repository. Null tracks the main branch
  - `variants` ContainerDeploymentTemplateVariantPublicApiDto[], required: Selectable weight variants. The first one is the default
    - `id` string, required: Variant identifier, passed as `variant` when deploying
    - `precision` string, required: Weight precision
    - `model_repository` string, required: Hugging Face repository this variant loads
    - `min_vram_gb` number, required: Minimum total GPU memory in GB
    - `is_default` boolean, required: Whether this variant is used when no variant is given
    - `description` string: What sets this variant apart
    - `recommended_compute` ComputeResource, required
      - `name` string, required: Name of the compute resource
      - `size` number, required: Number of compute units (e.g. 4 GPUs). Default is 1
  - `features` ContainerDeploymentTemplateFeaturePublicApiDto[], required: Independently selectable serving features
    - `id` string, required: Feature identifier, passed in `features` when deploying
    - `name` string, required: Human-readable name
    - `description` string: What the feature enables
    - `is_default` boolean, required: Whether the feature is on when no features are given
    - `conflicts_with` string[], required: Feature ids that cannot be enabled together with this one

## Other responses

- `401`: Unauthorized

## Changes

- **2026-09-14** `5d3a4de67175`: 1 info
  - endpoint added

[Operation history](https://skmtc.dev/datacrunch/apis/verda-cloud-public-api/changes/v1/container-deployment-templates/:template_id/get.md)

---

[API](https://skmtc.dev/datacrunch/apis/verda-cloud-public-api.md) · [All operations](https://skmtc.dev/datacrunch/apis/verda-cloud-public-api/llms.txt) · [OpenAPI document](https://skmtc.dev/datacrunch/apis/verda-cloud-public-api/revisions/05b167f797de?raw)