Pipecat: the OpenAI model id and GPT-6 tool calls

Where the model id lives

The OpenAI LLM page says OpenAILLMService provides chat completion. The settings table lists model as a string whose default is gpt-4.1. The basic setup example passes settings=OpenAILLMService.Settings(model="gpt-4o"). A later example uses model="gpt-4.1".

The constructor argument named model is marked deprecated since 0.0.105 and removed in 2.0.0. The page says to use settings=OpenAILLMService.Settings(...) instead.

If the API key argument is omitted, the page says the service uses the OPENAI_API_KEY environment variable. This page does not show a key.

from pipecat.services.openai.llm import OpenAILLMService

llm = OpenAILLMService(
    settings=OpenAILLMService.Settings(model="gpt-4.1"),
)

Source: Pipecat OpenAI LLM , checked October 25, 2026.

How to pin or swap it

Change the string in Settings(model=...). That is the OpenAI model id. The page does not document a router alias or a fallback list.

The documented default is the short id gpt-4.1. The shutdown table lists dated snapshot ids. Read the alias sentence under the table before you treat gpt-4.1 itself as shut down. If a dated row is the id you send, swap that string for the official replacement on the row.

Source: Pipecat OpenAI LLM , checked October 25, 2026.

GPT-6 function tools

The Responses page says to use OpenAILLMService for the Chat Completions API and OpenAIResponsesLLMService or OpenAIResponsesHttpLLMService for the Responses API. Both Responses classes take settings.model. Their settings table also defaults model to gpt-4.1.

Both the chat page and the Responses page say function calling is registered with register_function. The Responses page says the WebSocket variant is the one to use when the workflow involves frequent tool calls.

OpenAI's Using GPT-6 guide says GPT-6 Astra and GPT-6.1 Sol tool calling requires the Responses API. A pipeline that still constructs OpenAILLMService is still on Chat Completions, which is the endpoint that guide says cannot run those function tools. GPT-6 Sol and GPT-6 Luna are the models that guide allows on Chat Completions only with reasoning effort none. The error pages record the failure text.

from pipecat.services.openai.responses.llm import OpenAIResponsesLLMService

llm = OpenAIResponsesLLMService(
    settings=OpenAIResponsesLLMService.Settings(model="gpt-4.1"),
)

Sources: Pipecat OpenAI LLM , Pipecat OpenAI Responses LLM , Using GPT-6 , checked October 25, 2026.

Upcoming shutdowns in this dataset

gpt-4.1-2025-04-14, gpt-4.1-mini-2025-04-14, gpt-4o-2024-08-06, gpt-4o-2024-11-20, and gpt-4o-mini-2024-07-18 shut down on 2027-01-14. The official replacement on these rows is gpt-6-luna.

gpt-4.1, gpt-4.1-mini, gpt-4o, and gpt-4o-mini have no shutdown date on their own records. A request that sends the short id is not the dated snapshot row. Open the model page before you treat the short id as shut down.

Rows are upcoming OpenAI chat models in the gpt-4o, gpt-4.1, gpt-4-turbo, gpt-4, and gpt-3.5-turbo families. Audio, realtime, transcribe, and search ids are not in this table. The same id on another platform is in the other-platforms column. The full calendar is upcoming shutdowns. Provider wording for the shutdown day is on what happens on shutdown day.

Detect a dying model

These checks read SunsetDiff. They do not call the model provider.

• upcoming.json lists dated shutdowns in the next 90 days. No API key. Field notes are on the API docs.
• lookup/gpt-4o.json reads one id. Replace gpt-4o with the string in your config.
• The MCP server tools are check_model, upcoming_shutdowns, migration_diff, and scan_text.
• A status badge shows the stored status for a model page.
• The scan page builds a local search for retiring ids. Nothing is uploaded.

Questions