pluginsplugin-perplexity

ChatCompletion​Chat​Completion

yaml
type: "io.kestra.plugin.perplexity.ChatCompletion"

Ask a question to Perplexity

yaml
id: perplexity_chat
namespace: company.team

tasks:
  - id: ask_ai
    type: io.kestra.plugin.perplexity.ChatCompletion
    apiKey: '{{ secret("PERPLEXITY_API_KEY") }}'
    model: sonar
    messages:
      - type: USER
        content: "What is Kestra?"
    temperature: 0.7

Perplexity chat with Structured Output (JSON Schema)

yaml
id: perplexity_structured
namespace: company.name

tasks:
  - id: chat_completion_structured
    type: io.kestra.plugin.perplexity.ChatCompletion
    apiKey: '{{ secret("PERPLEXITY_API_KEY") }}'
    model: sonar
    messages:
      - type: USER
        content: "Make a JSON todo from this casual note: schedule team check-in next week; tags: work, planning;"
    jsonResponseSchema: |
      {
        "type": "object",
        "additionalProperties": false,
        "required": ["title", "done", "tags"],
        "properties": {
          "title": { "type": "string" },
          "done":  { "type": "boolean" },
          "tags":  { "type": "array", "items": { "type": "string" } },
          "notes": { "type": "string" }
        }
      }
Properties

API Key

The Perplexity API key used for authentication.

Messages

List of chat messages in conversational order.

Model

The Perplexity model to use (e.g., sonar, sonar-pro).

Default 0.0

Frequency Penalty

Decreases likelihood of repetition based on prior frequency. Valued between 0 and 2.0.

JSON Response Schema

JSON schema (as string) to force a custom Structured Output. If provided, the request will include response_format = { type: "json_schema", json_schema: { schema: } }.

The maximum number of tokens to generate.

Default 0.0

Presence Penalty

Positive values increase the likelihood of discussing new topics. Valued between 0 and 2.0.

Default false

Stream

Determines whether to stream the response incrementally.

Default 0.2

Temperature

The amount of randomness in the response, valued between 0 and 2.

Default 0

Top K

The number of tokens to keep for top-k filtering.

Default 0.9

Top P

The nucleus sampling threshold, valued between 0 and 1.

The generated text output

Full, raw response from the API.

Possible Values
SYSTEMASSISTANTUSER