Codex runs consistently into a stream error when using /compact
Resolved 💬 7 comments Opened Oct 11, 2025 by styler-ai Closed Nov 14, 2025
💡 Likely answer: A maintainer (etraut-openai, contributor)
responded on this thread — see the highlighted reply below.
What version of Codex is running?
• codex-cli 0.46.0
Which model were you using?
gpt-5-codex-high
What platform is your computer?
Windows 11
What steps can reproduce the bug?
/compact
stream error: unexpected status 400 Bad Request: {
"error": {
"message": "No tool call found for function call output with call_id call_9fuLHI3mxyYoaLpqTAYk2405.",
"type": "invalid_request_error",
"param": "input",
"code": null
}
What is the expected behavior?
That /compact compacts
What do you see instead?
stream error: unexpected status 400 Bad Request: {
"error": {
"message": "No tool call found for function call output with call_id call_9fuLHI3mxyYoaLpqTAYk2405.",
"type": "invalid_request_error",
"param": "input",
"code": null
}
Additional information
_No response_
7 Comments
0.46.0
Linux
The thing is, if you have a log running task you are jot always able to interrupt in time. /compact should always be posssible
Same problem with codex-cli 0.47.0 on MacOS Sequoia 15.6.1 (Intel Core i7)
Same problem here with codex-cli 0.47.0 on Windows 11 (WSL2 and native). gpt-5-codex-high. 10% context left, compact gives always
Same problem here info:
MacOS: 15.6.1
Chip: Apple M2 Pro
Installation Method: Homebrew
Error: unexpected status 400 Bad Request when using /compact
Auto compact fails as well.
Same here. The context got full, and when I typed /compact it also led to this. It did not auto compact for me. After the error message, it resets the context back to 100% but when I send a new prompt, it tells me the context is full.
Using WSL2 on Windows 11.
⚠️ stream error: unexpected status 400 Bad Request: {
"error": {
"message": "No tool call found for function call output with call_id call_FFqkhL94kEiJIcFoxGjuNGoL.",
"type": "invalid_request_error",
"param": "input",
"code": null
}
}; retrying 1/5 in 210ms…
⚠️ stream error: unexpected status 400 Bad Request: {
"error": {
"message": "No tool call found for function call output with call_id call_FFqkhL94kEiJIcFoxGjuNGoL.",
"type": "invalid_request_error",
"param": "input",
"code": null
}
}; retrying 2/5 in 375ms…
⚠️ stream error: unexpected status 400 Bad Request: {
"error": {
"message": "No tool call found for function call output with call_id call_FFqkhL94kEiJIcFoxGjuNGoL.",
"type": "invalid_request_error",
"param": "input",
"code": null
}
}; retrying 3/5 in 826ms…
⚠️ stream error: unexpected status 400 Bad Request: {
"error": {
"message": "No tool call found for function call output with call_id call_FFqkhL94kEiJIcFoxGjuNGoL.",
"type": "invalid_request_error",
"param": "input",
"code": null
}
}; retrying 4/5 in 1.566s…
⚠️ stream error: unexpected status 400 Bad Request: {
"error": {
"message": "No tool call found for function call output with call_id call_FFqkhL94kEiJIcFoxGjuNGoL.",
"type": "invalid_request_error",
"param": "input",
"code": null
}
}; retrying 5/5 in 3.43s…
■ unexpected status 400 Bad Request: {
"error": {
"message": "No tool call found for function call output with call_id call_FFqkhL94kEiJIcFoxGjuNGoL.",
"type": "invalid_request_error",
"param": "input",
"code": null
}
}
Closing in favor of #4813