Groq

OverviewCopy Anchor Link

Groq serves open models with fast inference on its LPUs.

• GroqLLMService: an OpenAILLMService subclass with Groq's base URL and default model.

InstallationCopy Anchor Link

bun add @pipedog/groq

It depends on @pipedog/openai, which is installed with it.

import { GroqLLMService } from "@pipedog/groq/llm";

PrerequisitesCopy Anchor Link

• A Groq account and an API key from the Groq console.
• Pass the key as apiKey. It is required, and the service never falls back to OPENAI_API_KEY. The examples read it from GROQ_API_KEY.

GroqLLMServiceCopy Anchor Link

ConfigurationCopy Anchor Link

The other options (organization, project, defaultHeaders, serviceTier, retryOnTimeout, retryTimeoutSecs, function calling) are OpenAILLMService's. See OpenAI.

SettingsCopy Anchor Link

Runtime-updatable settings, passed as settings: new GroqLLMService.Settings({...}) (GroqLLMSettings). They are OpenAI's settings, with these defaults:

Every other setting keeps OpenAI's default. Unset sampling settings are left out of the request, so the provider's own defaults apply. extra carries provider-specific request fields.

UsageCopy Anchor Link

import { LLMUpdateSettingsFrame } from "@pipedog/core/frames/frames";
import { GroqLLMService } from "@pipedog/groq/llm";

const llm = new GroqLLMService({
  apiKey: process.env.GROQ_API_KEY ?? "",
  settings: new GroqLLMService.Settings({
    model: "openai/gpt-oss-120b",
    systemInstruction: "You are a helpful assistant in a voice conversation.",
    temperature: 0.7,
  }),
});

Keep a Qwen model's reasoning out of the spoken reply with reasoningEffort: "none":

const qwen = new GroqLLMService({
  apiKey: process.env.GROQ_API_KEY ?? "",
  settings: new GroqLLMService.Settings({ model: "qwen/qwen3.8-27b", reasoningEffort: "none" }),
});

Change settings mid-call with an LLMUpdateSettingsFrame. The next completion uses them:

// Queue it on the worker, or push it from a processor upstream of the service.
const update = new LLMUpdateSettingsFrame({
  delta: new GroqLLMService.Settings({ temperature: 0.2, maxTokens: 300 }),
});

NotesCopy Anchor Link

• Runs a completion for each LLMContextFrame, between an LLMFullResponseStartFrame and an LLMFullResponseEndFrame, and streams the reply as LLMTextFrames, as OpenAILLMService does. Function calls, token usage metrics and runInference work the same way.
• "developer" messages are sent as is: the API supports that role.
• The same package has Groq's speech-to-text and text-to-speech services.

Runtime notesCopy Anchor Link

Generated from pipecat.runtimes and pipecat.compat in packages/groq/package.json: ✅ means the compat smoke test passes on that runtime.

• The client is the official openai JS SDK (through @pipedog/openai), which talks through fetch. It works on Bun, Node, Deno and Cloudflare Workers.
• The deprecated model constructor option is not ported. Use settings.
• Tested against a local OpenAI-compatible mock server, not against Groq itself.