Inline Assistant change default model

[rwese]

Check for existing issues

• [X] Completed

Describe the bug / provide steps to reproduce it

• have no access to the openai - gpt4 model
• modify user settings, or folder settings, to use the assistant > default_open_ai_model > 'gpt-3.5-turbo'
• attempt in-line assistant
• get error: Failed to connect to OpenAl API: The model 'gpt-4-0613 does not exist or you do not have access to it. ...

If I enable include conversation it will use the model and context of the chat, but often include the entire chat-response, when creating new content.

Environment

Zed: v0.102.2 (stable) OS: macOS 13.5.0 Memory: 16 GiB Architecture: aarch64

If applicable, add mockups / screenshots to help explain present your vision of the feature

even though the screenshot shows .zed/settings.json, where this is not supported per the documentation, I've set it also in the settings .config/zed/settings.json.

{
  "assistant": {
    "default_open_ai_model": "gpt-3.5-turbo"
  }
}

If applicable, attach your ~/Library/Logs/Zed/Zed.log file to this issue.

If you only need the most recent lines, you can run the zed: open log command palette action to see the last 1000.

No response

[dignifiedquire]

similar issue here, getting this for all models though

[barrynorman]

Same here. Unfortunately I cannot use OpenAI at all.

[otsulabs]

Same here. Unfortunately I cannot use OpenAI at all.

me too,