docs(vscode-lm): clarify VS Code LM API integration warning #8493
Conversation
…ording, and error guidance
dosubot
bot
added
the
size:XS
There was a problem hiding this comment.
Pull Request Overview
This PR clarifies the VS Code Language Model API integration warning to help users better understand potential model behavior differences and avoid common errors.
- Updates the description to use "provider" terminology instead of "vendor"
- Adds warning about potential behavior differences when models are wrapped or fine-tuned
- Provides specific instructions for proper model selection to avoid 400 errors
Tip: Customize your code reviews with copilot-instructions.md. Create the file or learn how to get started.
hannesrudolph
added
the
Issue/PR - Triage
![roomote[bot]](https://avatars.githubusercontent.com/in/1546624?s=60&v=4){kind=small}
There was a problem hiding this comment.
I found a localization regression: non-English locale files replaced localized copy with English. Recommend restoring localized strings in those files and limiting this change to the English locale, or follow up with proper translations. No other new issues found.
| "cacheUsageNote": "Hinweis: Wenn Sie keine Cache-Nutzung sehen, versuchen Sie ein anderes Modell auszuwählen und dann Ihr gewünschtes Modell erneut auszuwählen.", | ||
| "vscodeLmModel": "Sprachmodell", | ||
| "vscodeLmWarning": "Hinweis: Dies ist eine sehr experimentelle Integration und die Anbieterunterstützung variiert. Wenn Sie einen Fehler über ein nicht unterstütztes Modell erhalten, liegt das Problem auf Anbieterseite.", | ||
| "vscodeLmWarning": "Note: Models accessed via the VS Code Language Model API may be wrapped or fine-tuned by the provider, so behavior can differ from using the same model directly from a typical provider or router. To use a model from the Language Model dropdown, first switch to that model and then click Accept in the Copilot Chat prompt; otherwise you may see an error such as 400 \"The requested model is not supported\".", |
There was a problem hiding this comment.
P2: Localization regression — this replaces the previous German translation with English. Please keep the localized copy or provide a German translation to avoid degrading i18n completeness here and across other locales changed in this PR.
| "vscodeLmWarning": "Note: Models accessed via the VS Code Language Model API may be wrapped or fine-tuned by the provider, so behavior can differ from using the same model directly from a typical provider or router. To use a model from the Language Model dropdown, first switch to that model and then click Accept in the Copilot Chat prompt; otherwise you may see an error such as 400 \"The requested model is not supported\".", | |
| "vscodeLmWarning": "Hinweis: Dies ist eine sehr experimentelle Integration und die Anbieterunterstützung variiert. Wenn Sie einen Fehler über ein nicht unterstütztes Modell erhalten, liegt das Problem auf Anbieterseite.", |
| "cacheUsageNote": "提示：若未显示缓存使用情况，请切换模型后重新选择", | ||
| "vscodeLmModel": "VSCode LM 模型", | ||
| "vscodeLmWarning": "注意：这是一个非常实验性的集成，提供商支持会有所不同。如果您收到有关不支持模型的错误，则这是提供商方面的问题。", | ||
| "vscodeLmWarning": "Note: Models accessed via the VS Code Language Model API may be wrapped or fine-tuned by the provider, so behavior can differ from using the same model directly from a typical provider or router. To use a model from the Language Model dropdown, first switch to that model and then click Accept in the Copilot Chat prompt; otherwise you may see an error such as 400 \"The requested model is not supported\".", |
There was a problem hiding this comment.
P2: Localization regression — this replaces the previous Chinese translation with English. Please restore the localized string or provide a Chinese translation.
| "vscodeLmWarning": "Note: Models accessed via the VS Code Language Model API may be wrapped or fine-tuned by the provider, so behavior can differ from using the same model directly from a typical provider or router. To use a model from the Language Model dropdown, first switch to that model and then click Accept in the Copilot Chat prompt; otherwise you may see an error such as 400 \"The requested model is not supported\".", | |
| "vscodeLmWarning": "注意：这是一个非常实验性的集成，提供商支持会有所不同。如果您收到有关不支持模型的错误，则这是提供商方面的问题。", |
There was a problem hiding this comment.
I found some issues that need attention related to localization. The updated English warning text was copied into non-English locale files, replacing existing translations. This creates inconsistent UI and regresses i18n. Please keep the change in en only and either restore previous translations or mark them for translation.
| "cacheUsageNote": "Nota: Si no ve el uso del caché, intente seleccionar un modelo diferente y luego seleccionar nuevamente su modelo deseado.", | ||
| "vscodeLmModel": "Modelo de lenguaje", | ||
| "vscodeLmWarning": "Nota: Esta es una integración muy experimental y el soporte del proveedor variará. Si recibe un error sobre un modelo no compatible, es un problema del proveedor.", | ||
| "vscodeLmWarning": "Note: Models accessed via the VS Code Language Model API may be wrapped or fine-tuned by the provider, so behavior can differ from using the same model directly from a typical provider or router. To use a model from the Language Model dropdown, first switch to that model and then click Accept in the Copilot Chat prompt; otherwise you may see an error such as 400 \"The requested model is not supported\".", |
There was a problem hiding this comment.
[P2] The English warning text replaces the Spanish translation in this locale. Recommend reverting this line here (and in other non-English locales) to the previous localized copy and updating only en; track translation updates separately to avoid regressing i18n.
There was a problem hiding this comment.
I found some issues that need attention: the warning references a specific provider UI (Copilot Chat). This panel is provider-agnostic for the VS Code LM API, so referencing Copilot directly may confuse users of other providers. Recommend wording that generalizes the acceptance step while keeping Copilot as an example.
| "cacheUsageNote": "Note: If you don't see cache usage, try selecting a different model and then selecting your desired model again.", | ||
| "vscodeLmModel": "Language Model", | ||
| "vscodeLmWarning": "Note: This is a very experimental integration and provider support will vary. If you get an error about a model not being supported, that's an issue on the provider's end.", | ||
| "vscodeLmWarning": "Note: Models accessed via the VS Code Language Model API may be wrapped or fine-tuned by the provider, so behavior can differ from using the same model directly from a typical provider or router. To use a model from the Language Model dropdown, first switch to that model and then click Accept in the Copilot Chat prompt; otherwise you may see an error such as 400 \"The requested model is not supported\".", |
There was a problem hiding this comment.
[P2] Generalize provider-specific instruction: This warning references “Copilot Chat”, but the VS Code LM API can be provided by other extensions. Consider replacing 'click Accept in the Copilot Chat prompt' with 'click Accept in your provider's prompt (e.g., Copilot Chat)' to avoid provider-specific wording.
hannesrudolph
added
PR - Needs Review
and removed
Issue/PR - Triage
github-project-automation
bot
moved this from PR [Needs Review]
to Done
in Roo Code Roadmap
This PR now updates the correct warning in the settings UI:
New warning text:
"Note: Models accessed via the VS Code Language Model API may be wrapped or fine-tuned by the provider, so behavior can differ from using the same model directly from a typical provider or router. To use a model from the Language Model dropdown, first switch to that model and then click Accept in the Copilot Chat prompt; otherwise you may see an error such as 400 "The requested model is not supported"."