Fireworks AI

OverviewCopy Anchor Link

Fireworks AI serves open and proprietary models through an OpenAI-compatible API.

• FireworksLLMService: an OpenAILLMService subclass with Fireworks AI's base URL and default model.

InstallationCopy Anchor Link

bun add @pipedog/fireworks

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

import { FireworksLLMService } from "@pipedog/fireworks/llm";

PrerequisitesCopy Anchor Link

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

FireworksLLMServiceCopy 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 FireworksLLMService.Settings({...}) (FireworksLLMSettings). 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 { FireworksLLMService } from "@pipedog/fireworks/llm";

const llm = new FireworksLLMService({
  apiKey: process.env.FIREWORKS_API_KEY ?? "",
  settings: new FireworksLLMService.Settings({
    model: "accounts/fireworks/models/nemotron-3-ultra-nvfp4",
    systemInstruction: "You are a helpful assistant in a voice conversation.",
    temperature: 0.7,
  }),
});

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 FireworksLLMService.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.
• Model names use the accounts/fireworks/models/ prefix.
• seed, max_completion_tokens, stream_options and service_tier are not sent. Without stream_options, token usage is reported only when Fireworks includes it anyway.

Runtime notesCopy Anchor Link

Generated from pipecat.runtimes and pipecat.compat in packages/fireworks/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 Fireworks AI itself.