Codex Luna subagent fails with the default service tier
Nobody has claimed this yet.
- Dominant language
- Rust
- Stars
- 125k
- Forks
- 19.4k
- PR merge metrics
- PR metrics pending
Description
What version of the Codex App are you using (From “About Codex” dialog)?
Version 26.814.41407
What subscription do you have?
Enterprise API login
What platform is your computer?
Darwin 24.6.0 arm64 arm
What issue are you seeing?
When using GPT-5.6 Luna subagents, the native spawn_agent path rejects default, flex, and auto. The only accepted option is priority/Fast
What steps can reproduce the bug?
Request one Luna subagent in a Codex thread with model gpt-5.6-luna, reasoning effort to max and service tier as default. Prompt it to do something like say hello world. The request fails with: Service tier default is not supported for model gpt-5.6-luna. Supported service tiers: priority
What is the expected behavior?
The Luna subagent starts successfully and returns hello world.
Additional information
No response
Contributor guide
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
Research direction
Start at the native spawn_agent path for Luna subagents and reproduce the request with model gpt-5.6-luna, maximum reasoning effort, and service tier default. Trace how service tiers are passed to the subagent and verify that the request starts successfully and returns the expected hello-world response without rejecting default.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- rust
- Domain
- ai, cli
- Issue type
- Bug
- Difficulty
- 3/5
- Estimated time
- 1-2 days
- Activity status
- Active
- Clarity
- Mostly clear
- Newbie friendliness
- 65/100