Forked from lmstudio/openai-compat-endpoint
claude-sonnet-4-20250514claude-opus-4-20250514gpt-4.1-2025-04-14gpt-4.1-mini-2025-04-14deepseek-v4-flashdeepseek-v4-proglm-4.7-flashglm-5.2Use Custom ModelOpen the plugin settings in LM Studio and fill only the API key you need for the selected model.
| Model family | Required field | Default base URL |
|---|---|---|
claude-* | Anthropic API Key | https://api.anthropic.com/v1 |
gpt-* | OpenAI API Key | https://api.openai.com/v1 |
deepseek-* | Deepseek API Key | https://api.deepseek.com/v1 |
glm-* | GLM API Key | https://api.z.ai/api/paas/v4 |
Use Custom Model | Custom API Key | Set (Optional) Override Base URL |
Use Use Custom Model when the model is not in the preset list.
Required fields:
Custom Model: exact model id from the providerCustom API Key: API key for that provider(Optional) Override Base URL: required for custom providersExample:
Custom Model is always visible because LM Studio plugin config uses static fields. The plugin only reads it when Model is set to Use Custom Model.
Model: Use Custom Model
Custom Model: provider-model-id
Custom API Key: sk-...
(Optional) Override Base URL: https://api.example.com/v1