Groq integration

What it does

The Groq integration lets your agency invoke chat completions on Groq-hosted open models (Llama 3.x, Mixtral, and others) from inside any workflow. Groq's LPU inference returns tokens in a fraction of the wall-clock time of mainstream providers, so it is well suited to latency-sensitive client work such as real-time form triage, inline summarization of inbound messages, and high-volume classification. Connect a client's Groq key once and your workflows can swap between models, look up the active model catalog for a connected key, and inspect per-model context windows without leaving TaskJuice.

Connect a Groq account

1. Open your workspace in TaskJuice and navigate to Connections.
2. Choose Groq and click Connect.
3. In a new tab, open the Groq Console keys page signed in as the client (or as your agency, if the client has delegated key creation to you).
4. Click Create API Key, name it after the workspace, and copy the gsk_... value.
5. Paste the key into TaskJuice and save the connection.

To rotate or revoke the key later, return to the Groq Console keys page and delete the entry; create a new key and update the TaskJuice connection.

Triggers

Groq does not publish a public webhook surface, so the integration is action-only. Use a Schedule or another app's trigger to invoke Groq actions inside a workflow.

Actions

• groq/create-chat-completion generates a chat completion against a chosen Groq-hosted model (for example, llama-3.3-70b-versatile or llama-3.1-8b-instant) with a list of messages and optional temperature, max tokens, top-p, and response-format controls.
• groq/get-model returns metadata for a single Groq-hosted model by ID, including its context window and active status.
• groq/list-models returns the model catalog available to the authenticated key, useful for surfacing a model picker to end users or for branching on model availability.

Known limitations

• The integration calls the Groq API (an OpenAI-compatible surface at api.groq.com/openai/v1) with the connected key. Per-key rate limits, daily token caps, and model-access entitlements are governed by the client's Groq account tier, not by TaskJuice. When Groq returns a 429, TaskJuice surfaces it as a retryable rate-limit error and honors the recommended backoff.
• Streaming responses are not exposed. Each chat completion runs as a single request and resolves only after the full response is returned by Groq.
• Audio transcription and translation are not currently surfaced as first-class actions. If a client workflow needs Groq Whisper, reach out and we will scope it.
• The integration is action-only. To react to events in real time, use a Schedule trigger, a polling trigger from another app, or an inbound webhook from a different source and call Groq inside the workflow.