# Huggingface

Type: `conversation.huggingface`

Status: alpha

Reference: https://docs.dapr.io/reference/components-reference/supported-conversation/hugging-face/

## Example

```
apiVersion: cra.diagrid.io/v1beta1
kind: Component
metadata:
  name: <name>
spec:
  type: conversation.huggingface
  version: v1
  metadata:
    # API key for Huggingface.
    - name: key
      value: | 
        **********
    # A time-to-live value for a prompt cache to expire. Uses Golang durations (Optional)
    #- name: cacheTTL
    #  value: "10m"
    # Custom OpenAI-compatible endpoint URL for the model. If not specified, automatically generates the endpoint based on the model name using the template: https://router.huggingface.co/hf-inference/models/{{model}}/v1 (Optional)
    #- name: endpoint
    #  value: "https://router.huggingface.co/hf-inference/models/microsoft/DialoGPT-medium/v1"
    # The Huggingface model to use. Uses OpenAI-compatible API. Configurable via HUGGINGFACE_MODEL environment variable. (Optional)
    #- name: model
    #  value: "deepseek-ai/DeepSeek-R1-Distill-Qwen-32B"

```

## Authentication profiles

Available authentication profiles:

- API Key

### API Key

Authenticate using an API key

#### `key` (string)

**Required** - API key for Huggingface.

Example value:

```
**********
```

## Metadata

### `cacheTTL` (string)

A time-to-live value for a prompt cache to expire. Uses Golang durations

Example value: `10m`

### `endpoint` (string)

Custom OpenAI-compatible endpoint URL for the model. If not specified, automatically generates the endpoint based on the model name using the template: https://router.huggingface.co/hf-inference/models/\{\{model}}/v1

Example value: `https://router.huggingface.co/hf-inference/models/microsoft/DialoGPT-medium/v1`

### `model` (string)

The Huggingface model to use. Uses OpenAI-compatible API. Configurable via HUGGINGFACE\_MODEL environment variable.

Default value: `deepseek-ai/DeepSeek-R1-Distill-Qwen-32B`

Example value: `deepseek-ai/DeepSeek-R1-Distill-Qwen-32B`