Skip to content

Custom OpenAI Compatible model configuration not showing up in Model Selector via chatLanguageModels.json #289003

Description

Image Image Image

Describe the bug
I am trying to configure a custom OpenAI Compatible provider (GLM-4.7 via Zhipu/Custom endpoint) by editing the chatLanguageModels.json file. However, after saving the configuration and reloading the window, the custom model does not appear in the Language Models list, and I cannot select it.

To Reproduce

Open VS Code Insiders.
Navigate to User/chatLanguageModels.json.
Add the following configuration:
[
{
"name": "OpenAI Compatible",
"vendor": "customoai",
"family": "glm-4",
"version": "4.7",
"id": "glm-4.7",
"maxInputTokens": 204800,
"endpoint": "https://api.z.ai/api/coding/paas/v4",
"apiKey": "${input:chat.lm.secret.TOKEN}",
"model": "glm-4.7"
}
]

Save the file and run "Developer: Reload Window".
Open the Language Models tab or Chat interface.
Attempt to search for "OpenAI Compatible" or "glm-4.7".

Expected behavior
The custom model should appear in the selection list and be available for chat.

Metadata

Metadata

Labels

No labels
No labels

Type

No type

Projects

No projects

Milestone

No milestone

Relationships

None yet

Development

No branches or pull requests

Issue actions