Enable conversation-scoped speed selection for Cloud turns (#61178)

## What changed

- Stage Cloud speed choices for the next new turn in the conversation, alongside pending model and effort selections. Update the displayed conversation speed after admission without changing the local default.
- Offer speed commands supported by the pending model. Clear an incompatible pending speed change with a warning when switching models.
- Carry `service_tier` through Cloud message requests, preserving omission to inherit the thread tier and explicit null to clear it. Translate `default` to null and reject tier overrides on queued requests.

## Testing

Add tests for Cloud speed toggling, model-dependent speed choices, admission-time updates, and preserving the conversation tier when selecting a model. Extend request and wire tests to cover service-tier values, explicit clearing, and queue rejection.

GitOrigin-RevId: b4daa701f1fe29593d90a938e60cc20cb980298a
This commit is contained in:
Benjamin Carlsson
2026-11-12 20:56:33 +00:00
committed by copyberry
parent 72532ecae5
commit b279cfb170
20 changed files with 583 additions and 73 deletions
+28 -20
View File
@@ -103,36 +103,44 @@ impl App {
);
return;
};
if service_tier.is_some()
|| model == crate::model_catalog::LUNA_RESERVE_MODEL
|| !self.model_catalog.models.iter().any(|preset| {
preset.model == model
&& effort.as_ref().is_none_or(|effort| {
*effort == preset.default_reasoning_effort
|| preset
.supported_reasoning_efforts
.iter()
.any(|preset| &preset.effort == effort)
})
&& preset.model_specialty.as_deref() != Some(MODEL_SPECIALTY_CYBER)
})
{
let preset = self.model_catalog.models.iter().find(|preset| {
preset.model == model
&& effort.as_ref().is_none_or(|effort| {
*effort == preset.default_reasoning_effort
|| preset
.supported_reasoning_efforts
.iter()
.any(|preset| &preset.effort == effort)
})
&& preset.model_specialty.as_deref() != Some(MODEL_SPECIALTY_CYBER)
});
let Some(preset) = preset.filter(|_| {
service_tier.is_none() && model != crate::model_catalog::LUNA_RESERVE_MODEL
}) else {
self.chat_widget.add_error_message(
"Choose a supported Cloud model and effort at the current speed.".into(),
);
return;
}
};
let label = Self::reasoning_label_for(&model, effort.as_ref())
.map_or_else(String::new, |label| format!(" {label}"));
let message =
format!("Selected {model}{label} for the next new turn in this conversation");
if app_server.stage_cloud_model_selection(thread_id, model, effort) {
self.chat_widget.add_info_message(message, /*hint*/ None);
} else {
self.chat_widget.add_error_message(
match app_server.stage_cloud_model_selection(thread_id, preset, effort) {
Ok(dropped_tier) => {
self.chat_widget.sync_service_tier_commands();
self.chat_widget.add_info_message(message, /*hint*/ None);
if let Some(tier) = dropped_tier {
let tier = if tier == "priority" { "fast" } else { &tier };
self.chat_widget.add_warning_message(format!(
"{model} does not support {tier} speed. Cleared the pending speed change."
));
}
}
Err(()) => self.chat_widget.add_error_message(
"The Cloud account changed. Reopen the conversation before choosing a model."
.into(),
);
),
}
return;
}