Skip to content
CC Switch WikiUnofficial Community Wiki

Changed files cannot show diff below. Codex

Quick fix

Do not automatically fetch 火山方舟 Coding Plan models. Manually use deepseek-v4-flash, then restart Codex and CC Switch.

Symptom

Section titled “Symptom”
Error output
改动文件无法在下方看到diff。
English translation

Changed files cannot show the diff below.

Cause

Section titled “Cause”

The report states that CC Switch’s “fetch model list” request to `/api/coding/v1/models` returns pay-as-you-go Model IDs from the 方舟 model marketplace, such as `deepseek-v4-flash-ga-260731`, which are unsuitable for Coding Plan. The manual support guidance in the report recommends entering a Coding Plan model ID without the date suffix.

The original post provides no maintainer confirmation or resolved status; its configuration method comes from a 火山方舟 support reply. Comments also report that many automatically fetched 火山方舟 Coding Plan model names are incorrect.

Fix

Section titled “Fix”

  1. Do not use “fetch model list”; manually add a model in CC Switch model configuration and enter `deepseek-v4-flash` as the model ID.

  2. Confirm the provider request URL is `https://ark.cn-beijing.volces.com/api/coding/v3`. If “full URL” is selected, enable the routing master switch in CC Switch under “Settings → Routing → Local Routing” and select the tool corresponding to Codex.

  3. Change the Codex model to `deepseek-v4-flash` and keep `wire_api = "responses"`.

    Codex 的 config.toml
    model = "deepseek-v4-flash"
    wire_api = "responses"

  4. Fully quit Codex and CC Switch, then reopen them and check whether the change card appears.

Affected Versions

Section titled “Affected Versions”
ToolCodex
Version受影响版本：v3.20.0
PlatformsWindows

Source Issues

Section titled “Source Issues”

This page is distilled from 1 real issues

FAQ

Section titled “FAQ”
Are many automatically fetched 火山方舟 Coding Plan model names incorrect?
Comments on this issue report that many automatically fetched 火山方舟 Coding Plan model names are incorrect; the original post recommends avoiding automatic fetching and manually entering `deepseek-v4-flash`.
Section titled “Related problems”
Codex desktop cannot see modelsSign in to official ChatGPT, enable “Keep official login when switching to third parties,” and enable local routing. Restart Codex.6 issues mergedCodex shows no model list and will not switch modelsTurn on keeping the official login when switching to a third party in cc-switch, and make sure you signed in to official ChatGPT in Codex once.6 issues mergedGateway rejected model "claude-haiku-4-5"Enable routing based on the API type; for 火山方舟 Responses API, use the full URL and set `isFullUrl=true` to prevent the path from being appended again.2 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.