Easy way to fix Continue Edit Models put its reasoning content into code.

[Z2IRIM]

hey, I just found how to fix this. Edit model works fine again when I disabled reasoning on reasoningable model like gpt5.4. If you have a openai-compatible model, edit your config.yaml like:

• name: gpt-5.4-mini
  provider: openai
  model: gpt-5.4-mini
  apiKey: x
  apiBase: x
  roles:
  • edit
    requestOptions:
    extraBodyProperties:
    reasoning:
    effort: none

---

by reasoning:effort none (follow format above), you can turn off reasoning, which is cleared by OpenAI docs. If you are using another provider, check docs and search how to disable it. After disabling this, the response from model is no longer with thinking process, just the code only like below:
---------Log-from-Continue-Console--------------------------------

{isDark ? (

) : (

)}

);
}

Now the response is very clean, and apply correctly.