Cerebras

OverviewCopy Anchor Link

Cerebras serves open models with fast inference on its own hardware.

• CerebrasLLMService: an OpenAILLMService subclass with Cerebras's base URL and default model.

InstallationCopy Anchor Link

bun add @pipedog/cerebras

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

import { CerebrasLLMService } from "@pipedog/cerebras/llm";

PrerequisitesCopy Anchor Link

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

CerebrasLLMServiceCopy 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 CerebrasLLMService.Settings({...}) (CerebrasLLMSettings). 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";

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

Cerebras-only request fields, such as reasoning_effort, go in extra:

const reasoning = new CerebrasLLMService({
  apiKey: process.env.CEREBRAS_API_KEY ?? "",
  settings: new CerebrasLLMService.Settings({ extra: { reasoning_effort: "low" } }),
});

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 CerebrasLLMService.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.
• Cerebras honors max_tokens, max_completion_tokens and the penalties, so they are sent when set.

Runtime notesCopy Anchor Link

Generated from pipecat.runtimes and pipecat.compat in packages/cerebras/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 Cerebras itself.