refactor: group assistant runtime controls
This commit is contained in:
1 parent
ebce24ba67
commit
68d236fd63
3 files changed
+192
-64
No files matched your search
@@ -95,11 +95,11 @@ as part of the prompt.
|
||||
|
||||
Type `@` in the composer to add permission-checked context. The nested menu contains apps, integrations,
|
||||
boards, widgets, and tools available to the current user. Homarr validates every referenced ID again on
|
||||
the server before adding it to the model context. Type `/` for common prompt shortcuts. The compact
|
||||
controls below the composer can switch between models discovered from the configured endpoint. Model
|
||||
selection is stored on the conversation after a completed request. Thinking mode can be left on
|
||||
**Auto**, disabled, or set from minimal through high; explicit levels are sent using the AI SDK's
|
||||
provider-neutral reasoning setting and still depend on support from the chosen model and endpoint.
|
||||
the server before adding it to the model context. Type `/` for common prompt shortcuts. The joined
|
||||
controls below the composer open a searchable model picker and the thinking-level menu. Model selection
|
||||
is stored on the conversation after a completed request. Thinking mode can be left on **Auto**, disabled,
|
||||
or set from minimal through high; explicit levels are sent using the AI SDK's provider-neutral reasoning
|
||||
setting and still depend on support from the chosen model and endpoint.
|
||||
|
||||
Each completed response includes a circular context-window meter. Select it to open request details when
|
||||
the provider reports them: input, output, cached-input and reasoning tokens, used and remaining context,
|
||||
|
||||
Reference in new issue
Block a user