docs/integrations/chat/openai/

[giscus[bot]]

docs/integrations/chat/openai/

This notebook provides a quick overview for getting started with OpenAI chat models. For detailed documentation of all ChatOpenAI features and configurations head to the API reference.

https://python.langchain.com/docs/integrations/chat/openai/

[syarif1977]

berikan contoh pembangunan infrastruktur

[saocorley]

Hello,
Is there a way of streaming audo using openai-audio-preview and langraph? It seems that the responses I get from that are empty. Any ideas?

[jorgehermo9]

I have the same problem. Is there an open issue about it?

[droidnoob]

I have added an issue #29776

[Erybor]

Does Langchain support the new GPT o1 / o1-mini models? Trying to pass model to the ChatOpenAI returns error: "Unsupported parameter: 'temperature' is not supported with this model."

[timonschramm]

llm = ChatOpenAI(
temperature=1,
model="o1-mini",
streaming=True,
openai_api_key=st.secrets["OPENAI_API_KEY"]
)

try setting it to 1, it worked for me

[BautistaGuerra]

Could you use tool_calling with 'o1-mini'?

[BautistaGuerra]

Is the model 'gpt-4o-2024-11-20' available to use with ChatOpenAI? And the same for 'o1'.
I've upgraded langchain but I'm still encountering this error:
'The model o1 does not exist or you do not have access to it.'

[Huyueeer]

I had the same problem

[eladrave]

AFAIK, o1 does not support tool or function calling

[Akinola92]

Up until a few days ago, I was able to run the line "from langchain_openai import ChatOpenAI", but now I'm receiving the following error message "ImportError: cannot import name 'convert_to_openai_image_block' from 'langchain_core.messages' (/usr/local/lib/python3.11/dist-packages/langchain_core/messages/init.py)".

Any ideas on what's going wrong and how to fix it?

[itsphily]

I'm running langchain-openai v0.3.16 and I have some trouble with the "use_responses_api": True parameter.

I tried setting both inside and outside kwargs and it didn't work
llm_long_context_tool_use = ChatOpenAI(model="o4-mini",
temperature=0,
timeout=None,
max_retries=4,
model_kwargs={"reasoning_effort": "high",
"use_responses_api": True}
)

llm_long_context_tool_use = ChatOpenAI(model="o4-mini",
temperature=0,
use_responses_api=True,
timeout=None,
max_retries=4,
model_kwargs={"reasoning_effort": "high"}
)

[Sauloopez]

Is possible to work with OpenAI threads for chat context?

I need integrate the openai threads and response streams to handle chat context.

[kv-22]

Is there a way to use the built-in tools with a tool I've defined as a function?