Skip to content
CC Switch WikiUnofficial Community Wiki

Not supported model gpt-5.5: Cline automatic failover failed

Quick fix

Upgrade to the latest CC Switch 4.0 release and retry. If it still fails, provide the version, system, reproduction steps, and logs.

Symptom

Section titled “Symptom”
Error output
{"message":"CC Switch local proxy failed while handling Codex endpoint /chat/completions. Provider: xiaomi; model: gpt-5.5; upstream_status: HTTP 400; cause: Param Incorrect","type":"upstream_error","code":"400","param":"Not supported model gpt-5.5","provider":"xiaomi","model":"gpt-5.5","endpoint":"/chat/completions","upstream_status":400}}

Cause

Section titled “Cause”

The same failover configuration could switch to 小米 mimo in the VS Code Codex extension, but when the Cline extension requested `/chat/completions`, the upstream received the mapped `gpt-5.5` and returned HTTP 400 because the model was unsupported. The reporter noted that they might use the Responses and completions APIs, respectively.

Another comment described an HTTP 400 from a `/responses` request exceeding the context length after cross-model compaction, but this was a different error variant in the same discussion. The issue was closed as timed out because no new reproduction details were provided. The root cause was not confirmed, and no successful fix was verified. The maintainer only said that 4.0 substantially refactored the related functionality and recommended updating and retrying.

Fix

Section titled “Fix”

  1. Update CC Switch to the latest 4.0 release, then retest the Cline request using the existing local proxy address and failover configuration.

  2. If the problem persists, resubmit the CC Switch version, operating system, complete reproduction steps, and relevant logs or screenshots. The current discussion provides no confirmed configuration change that resolves it.

Affected Versions

Section titled “Affected Versions”
ToolCline
VersionUnknown
PlatformsUnknown

Source Issues

Section titled “Source Issues”

This page is distilled from 1 real issues

FAQ

Section titled “FAQ”
Why does the same automatic failover work in the Codex extension but return `Not supported model gpt-5.5` in the Cline extension?
The reporter observed that Codex uses Responses while Cline uses completions. Cline's `/chat/completions` request sent the mapped `gpt-5.5` to the 小米 upstream, which rejected it with HTTP 400. The discussion did not confirm a more specific root cause.
What information should I provide if it still fails after upgrading?
The maintainer requested the CC Switch version, system, complete reproduction steps, and relevant logs or screenshots before reopening the issue.
Section titled “Related problems”
got 0 from 0 output items: Codex remote compaction failedUpgrade CC Switch to v4.0.0 or later. If you cannot upgrade yet, disable Remote Compaction and use Codex local compaction.4 issues mergedmaximum context length is 1048565 tokens.Reduce PDF or message input. If the DeepSeek-compatible endpoint does not support PDF, stop uploading it, update CC Switch, and retry.1 issues mergedunexpected status 404 Not FoundUpgrade cc-switch to v3.16.4+ and set the upstream format explicitly to Chat Completions on the Codex provider form.4 issues merged

This is an unofficial community wiki with no affiliation to the cc-switch authors or the project itself. Its content is compiled from the project's public GitHub issues. This site distributes no software.