`service_tier = "flex"` is omitted for `gpt-5.5` even though docs say it is supported

What version of Codex CLI is running?

0.143.0 - 0.144.4

What subscription do you have?

Pro

Which model were you using?

gpt-5.5, gpt-5.6

What platform is your computer?

Linux 5.15.209 x86_64 Intel(R) Xeon(R) E-2146G CPU @ 3.50GHz

What terminal emulator and version are you using (if applicable)?

_No response_

Codex doctor report

What issue are you seeing?

Observed behaviour

Codex emits this warning:

Configured service tier flex is not advertised as supported for model gpt-5.5 and will be omitted from requests.

The local model cache fetched by Codex lists gpt-5.5 as API-supported, but only advertises the priority service tier:

  {
    "slug": "gpt-5.5",
    "supported_in_api": true,
    "additional_speed_tiers": ["fast"],
    "service_tiers": [
      {
        "id": "priority",
        "name": "Fast",
        "description": "1.5x speed, increased usage"
      }
    ]
  }

Cache metadata:

  {
    "fetched_at": "2026-07-08T10:36:44.186658278Z",
    "client_version": "0.143.0"
  }

Documentation references

• Codex config reference documents service_tier and includes flex: https://developers.openai.com/codex/config-reference
• Flex Processing docs show gpt-5.5 with service_tier: "flex": https://developers.openai.com/api/docs/guides/flex-processing

What steps can reproduce the bug?

Config

service_tier = "flex"
model = "gpt-5.5"
model_reasoning_effort = "medium"

What is the expected behavior?

service_tier = "flex" should be sent for gpt-5.5, or the documentation/model catalogue should agree that Flex is unavailable for this model.

Additional information

Possible cause

The Codex CLI model catalogue for gpt-5.5 appears stale or incomplete: it advertises priority/Fast but not flex, so the CLI defensively strips service_tier = "flex" before making requests.