Conversation
…ith tests This gets rid of the awkward AGENT_MODEL constant by making the LLM an inherent property of the Agent, which is more intuitive. Also switched the test judge model to base 5.2 instead of the chat version to make it clearer that you can (and should) use a different model for evals than for core chat.
bcherry
changed the title
The constructor requires an options arg and destructures it; calling new voice.AgentSession() with no arg throws at runtime.
…sharing-node-tYly8
…e-tYly8' into claude/cleaner-model-sharing-node-tYly8
u9g
reviewed
May 4, 2026
openai 6.36.0 changed its credential check to reject empty-string apiKey, but @livekit/agents@1.3.x hardcodes apiKey: '' when constructing its OpenAI client inside inference.LLM. Pin to <6.36.0 until the agents SDK is patched.
u9g
reviewed
May 7, 2026
u9g
approved these changes
May 7, 2026
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
this gets rid of the awkward AGENT_MODEL constant by just making the LLM an inherent property of the Assistant, which seems more intuitive
I also switched the test judge model to base 5.2 instead of the chat version to make it clearer that you can (and should) use a different model for evals than for core chat.
I'd like to move STT and TTS as well, but I found a bug we need to fix first
also see livekit-examples/agent-starter-python#71