-
-
Notifications
You must be signed in to change notification settings - Fork 79.2k
[Bug]: using codex harness with gpt-5.4-nano causes errors #85806
Copy link
Copy link
Closed
Labels
P2Normal backlog priority with limited blast radius.Normal backlog priority with limited blast radius.bugSomething isn't workingSomething isn't workingclawsweeper:fix-shape-clearClawSweeper found a clear likely implementation shape for this issue.ClawSweeper found a clear likely implementation shape for this issue.clawsweeper:queueable-fixClawSweeper marked this issue as an existing queue_fix_pr work candidate.ClawSweeper marked this issue as an existing queue_fix_pr work candidate.clawsweeper:source-reproClawSweeper found a high-confidence source-level issue reproduction.ClawSweeper found a high-confidence source-level issue reproduction.impact:auth-providerAuth, provider routing, model choice, or SecretRef resolution may break.Auth, provider routing, model choice, or SecretRef resolution may break.issue-rating: 🦞 diamond lobsterVery strong issue quality with high-confidence source-level or clear reproduction.Very strong issue quality with high-confidence source-level or clear reproduction.
Metadata
Metadata
Assignees
Labels
P2Normal backlog priority with limited blast radius.Normal backlog priority with limited blast radius.bugSomething isn't workingSomething isn't workingclawsweeper:fix-shape-clearClawSweeper found a clear likely implementation shape for this issue.ClawSweeper found a clear likely implementation shape for this issue.clawsweeper:queueable-fixClawSweeper marked this issue as an existing queue_fix_pr work candidate.ClawSweeper marked this issue as an existing queue_fix_pr work candidate.clawsweeper:source-reproClawSweeper found a high-confidence source-level issue reproduction.ClawSweeper found a high-confidence source-level issue reproduction.impact:auth-providerAuth, provider routing, model choice, or SecretRef resolution may break.Auth, provider routing, model choice, or SecretRef resolution may break.issue-rating: 🦞 diamond lobsterVery strong issue quality with high-confidence source-level or clear reproduction.Very strong issue quality with high-confidence source-level or clear reproduction.
Type
Fields
Give feedbackNo fields configured for issues without a type.
Bug type
Regression (worked before, now fails)
Beta release blocker
No
Summary
(Not sure if this is actually a "regression" or not as I'm comparing to a version of openclaw that didn't include codex)
OC 2026.5.20 using codex and openai's gpt-5.4-nano results in errors and fallbacks
Steps to reproduce
start with 4.23 configured to use API and gpt-5.4-nano (Default) and gpt-5.4-mini (fallback)
Chat works fine.
Upgrade to 2026.5.20 and run "doctor --fix" a couple times to get the codex plugin going.
Ensure codex harness is in use and type anything in webchat while the model is gpt-5.4-nano.
Expected behavior
It should just work.
Actual behavior
OC falls back to gpt-5.4-mini and errors are written to the log (see below)
OpenClaw version
2026.5.20
Operating system
linux (debian/trixie)
Install method
install.sh followed by "openclaw update" (npm, I think)
Model
openai/gpt-5.4-nano
Provider / routing chain
direct to openai
Additional provider/model setup details
No response
Logs, screenshots, and evidence
Impact and severity
Affected: unknown
Severity: For those using nano to save money, the constant fallback to a more expensive model is severe, though changing to the pi harness works around the issue.
Frequency: 100% reproducible
Consequence: It really makes openAI's codex harness look bad.
Additional information
It is possible to switch to the pi model to work around the issue.