Skip to content
Recued
Menu
← Back to recipes

Groq chat completion brief

by recued-core v1

Low-latency Groq chat workflow. It reads the selected model metadata, creates one non-streaming chat completion from an explicit question, and summarizes answer quality, finish reason, model, and token usage. It never enables streaming, tools, tool_choice, functions, MCP tools, Compound custom tool configuration, web-search settings, audio, files, or arbitrary Groq routes.

How it works 7 steps

Inspect the data fetches, transforms, gates, and output this recipe runs.

Process (7 steps)
model_details ?
completion ?
choices default
Apply default
choice_count count
Count items in choices
summarize ?
choice_table to_table
Format results as a data table
card to_summary
Format results as a summary card
Settings 8 configurable

Configurable at install. Defaults shown — change them anytime in Recued.

groq setting =
focus setting = Summarize the Groq answer, model metadata, finish reason, uncertainty, and token usage. Use only the Groq response and model metadata.
model setting = llama-3.3-70b-versatile
question setting =
max length setting = 800
temperature setting = 0.2
system prompt setting = Answer concisely, call out uncertainty, and avoid inventing facts not present in the prompt.
max completion tokens setting = 900

Trust & control

What installing this recipe would let it do. Recued grants these permissions at install — review them there before approving.

Where it runs

Device + server Runs on your device (browser) or your server.

Permissions it requires

Read your Groq connection
Declared by the recipe — Recued grants these at install, where you review them before approving.

About

Tags

groq groqchatllminferenceanswer-reviewai-summarypack:groq

Details

7 steps 8 configurable settings recipe_id: groq-chat-completion-brief