ISSUE DESCRIPTION Add reasoning.effort support for OpenAI gpt-realtime-2 Feature Type…
- User
ISSUE DESCRIPTION
Add reasoning.effort support for OpenAI gpt-realtime-2
Feature Type
Would make my life easier
Feature Description
OpenAI
gpt-realtime-2supports configurable reasoning viareasoning.effort, but the Python OpenAI Realtime plugin does not seem to expose this option onopenai.realtime.RealtimeModel.It would be useful to support something like:
from livekit.plugins import openai llm = openai.realtime.RealtimeModel( model="gpt-realtime-2", reasoning={"effort": "low"}, )Workarounds / Alternatives
I don't think there is a clean workaround through the current Python
RealtimeModelAPI.The only workaround I can see is patching the plugin locally to add
reasoningto the session update payload.Additional Context
OpenAI docs list these effort values:
minimal,low,medium,high,xhigh.References:
- https://developers.openai.com/api/docs/models/gpt-realtime-2
- https://developers.openai.com/api/docs/guides/realtime-models-prompting#set-reasoning-effort
- Related JS issue: https://github.com/livekit/agents-js/issues/1433
- Related JS PR: https://github.com/livekit/agents-js/pull/1575
Interface
No new interfaces are introduced.
ADDITIONAL ADVICE
Since you are given a git repository, you can use git commands to simplify your work. For example, if you made a mistake and want to revert your changes, you can use
git checkout <file>to restore the file to its original state. You can also reset all changes in the repository usinggit reset --hardcommand. Additionally, you can usegit stashandgit stash popto temporarily save your changes and restore them later. Finally, you can see the changes that you've made (which will be submitted when you callsubmit) usinggit statusorgit diffcommands. However you don't need to usegit addorgit commitbefore submitting your changes. In fact,submitonly submits currently unstaged changes because it usesgit diffto get the changes that need to be submitted.CHECKLIST
Before submitting your solution, please go over the following checklist and make sure you've done everything:
- If an error was reported in the issue description, I have successfully reproduced it.
- If an error was reported in the issue description, I have confirmed that my fix resolves the error.
- I have thought about edge cases and made sure my fix handles them as well.
- I have run existing tests in the repository that might have been affected by the change I made and confirmed that they pass. I want you to list every bullet from this checklist and write your reasoning for why you think you did it or didn't need to.
Repository has been uploaded and your shell is currently at the repository root. Time to solve the issue!
(Current directory: /agents, current file: none) bash-$
- Assistant
I'll explore the repository to find the OpenAI Realtime plugin and how session options are currently passed.
- Assistant
I'll inspect the OpenAI realtime plugin and how other options are passed through.
- Assistant
rg is not available, let me use grep instead.
- Assistant
I'll open the RealtimeModel constructor and session-update logic next.
- Assistant
Let me look at RealtimeSessionCreateRequest for reasoning field and the RealtimeModel _RealtimeOptions and constructor.
- Assistant
I'll inspect how options flow into session updates so we can mirror that for reasoning.
- Assistant
I'll keep reading
update_optionsand check models/docs forgpt-realtime-2. - Assistant
Checking for a model-level
update_optionsand any related tests. - Assistant
I'll read model-level
update_optionsand Azure session mapping next. - Assistant
I'll check how other LLM code handles the
Reasoningtype for consistency. - Assistant
Using the venv Python to inspect types I need.
- Assistant
Checking how session options are copied from the model:
- Assistant
Implementing
reasoningsupport by mirroring thetruncationpattern. - Assistant
I need to open the file at line 87 and then edit.
- Assistant
I'll apply the import and options-dataclass changes first.
- Assistant
Adding a helper and updating
_RealtimeOptionsnext. - Assistant
Now update _RealtimeOptions
- Assistant
Need to navigate to that line first.
- Assistant
Now I can replace.
- Assistant
Updating both overloads and the main
__init__signature. - Assistant
The replace tool needs the lines in view; I'll use exact line edits instead.
- Assistant
The last edit corrupted the overload — fixing that section carefully.
- Assistant
First overload looks good. Updating the Azure overload and main
__init__next. - Assistant
Now update main init signature, docstring, and options construction.
- Assistant
Update docstring and opts construction.
- Assistant
Now find the _opts construction and add reasoning.
- Assistant
I'll add reasoning to opts construction.
- Assistant
Updating RealtimeModel.update_options next.
- Assistant
Update update_options on RealtimeModel
- Assistant
Updating session creation and RealtimeSession.update_options.
- Assistant
Let me goto those lines and edit.
- Assistant
Let me edit the session create and update_options for session.
- Assistant
There's a duplicate return line — fixing it.
- Assistant
There's still an extra
)on line 1219 — removing it. - Assistant
I'll restore the blank line before
@propertyand update sessionupdate_options. - Assistant
Now add the reasoning handling after truncation in session update_options.
- Assistant
Removing the duplicate
has_audio_configline. - Assistant
Checking how events are serialized so
reasoningisn't dropped: - Assistant
I'll write a short script to verify the session update payload includes reasoning.
- Assistant
I'll also verify the default path without reasoning.
- Assistant
I'll run a syntax check and ruff on the edited file.
- Assistant
Fixing import order with ruff.
- Assistant
Ruff fixed the imports; verifying the final diff next.
Other models, same or similar prompt
Responses were produced in different, uncontrolled contexts (system prompts, settings, prior turns). This is not a benchmark.
Identical first prompt
claude-sonnet-5Compare side by side →
claude-sonnet-5Compare side by side →
claude-sonnet-5Compare side by side →
claude-sonnet-5Compare side by side →
claude-sonnet-5Compare side by side →
Qwen/Qwen3.6-35B-A3BCompare side by side →
Qwen/Qwen3.6-35B-A3BCompare side by side →
Qwen/Qwen3.6-35B-A3BCompare side by side →
Qwen/Qwen3.6-35B-A3BCompare side by side →
Qwen/Qwen3.6-35B-A3BCompare side by side →
Qwen/Qwen3.6-27BCompare side by side →
Qwen/Qwen3.5-35B-A3BCompare side by side →