feat: model picker favorites, recents, and effort-aware selection - #6
Open
coda-rho-bot wants to merge 2 commits into
Open
feat: model picker favorites, recents, and effort-aware selection#6coda-rho-bot wants to merge 2 commits into
coda-rho-bot wants to merge 2 commits into
Conversation
…g for bottom sheets The app is edge-to-edge (targetSdk 36), so Android never resizes the window for the IME and @gorhom/bottom-sheet's built-in keyboardBehavior stays a no-op — sheets with search inputs sat under the keyboard. Listen to Keyboard events directly and grow the sheet's bottom padding by the keyboard height while visible; dynamic sizing re-measures and the whole sheet rides above it. Also: the Android hardware back button now dismisses the presented sheet instead of popping the screen underneath (BackHandler while a sheet is open, tracked via onChange), and an optional onSheetDismiss callback lets callers reset transient state after any dismissal path (back, drag, backdrop). Co-authored-by: Letta Code <noreply@letta.com>
Favorites (star toggle, persisted in AsyncStorage) and recents (last five used, per handle) sections above the full catalog, so the models you actually use are one tap away. The reasoning-effort segment now renders only for models that actually ship effort variants (distinct reasoning_effort entries across catalog entries sharing a handle) — models without variants show no segment. Tapping a model with variants keeps the sheet open so the tier can be picked in the same interaction; models without variants apply immediately. api.ts: widen ReasoningEffort with "max" (the catalog ships max variants and Anthropic accepts it; the SDK union omits it), surface each catalog entry's declared effort on ModelOption, and pass "max" through to the OpenAI reasoning settings. Stacked on the bottom-sheet fixes PR — uses the sheet's onSheetDismiss to clear the pending-effort state when the sheet closes. Co-authored-by: Letta Code <noreply@letta.com>
This was referenced Sep 1, 2026
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Makes the model picker faster to use:
Favorites + recents. Star any model (persisted in AsyncStorage) and the starred section pins to the top; the last five used models (keyed by handle) appear beneath it. The full catalog follows with search.
Effort-aware rendering. The reasoning-effort segment now renders only for models that actually ship effort variants — detected as distinct
reasoning_effortvalues across catalog entries sharing a handle. Models without variants (most of the catalog) show no dead segment. Tapping a model with variants keeps the sheet open so the tier can be picked in the same interaction; models without variants apply immediately.API groundwork: widens
ReasoningEffortwith "max" (the catalog ships max variants — claude and gpt-5.6 lines — and Anthropic's model settings accept it; the SDK union omits it), surfaces each catalog entry's declared effort onModelOption, and passesmaxthrough to OpenAI reasoning settings.Note: stacked on #5 — it uses the sheet's
onSheetDismissto clear pending effort state. Merge that first; this branch is based on it.AI Contribution Disclosure