Struct ModelSwitchToRequest 

pub struct ModelSwitchToRequest {

    pub auto_tier: Option<AutoTier>,
    pub compaction_decision: Option<String>,
    pub context_tier: Option<ContextTier>,
    pub defer_if_model_change_queued: Option<bool>,
    pub model_capabilities: Option<ModelCapabilitiesOverride>,
    pub model_change_scope: Option<String>,
    pub model_id: String,
    pub picker_persistence: Option<ModelPickerPersistenceRequest>,
    pub reasoning_effort: Option<String>,
    pub reasoning_summary: Option<ReasoningSummary>,
    pub repo_scope: Option<String>,
    pub require_available: Option<bool>,
    pub run_compaction_preflight: Option<bool>,
    pub source: Option<ModelChangeSource>,
    pub verbosity: Option<Verbosity>,
}

Target model identifier and optional reasoning effort, summary, capability overrides, and context tier.

Experimental. This type is part of an experimental wire-protocol surface and may change or be removed in future SDK or CLI releases.

Fields

auto_tier: Option<AutoTier>

Optional Auto routing preference to stage atomically with selecting auto. Pass null to return to provider-default Auto routing. This field is rejected when modelId is not auto.

compaction_decision: Option<String>

Explicit response to a model-switch compaction preflight. Omit to request a confirmation projection when compaction is necessary.

context_tier: Option<ContextTier>

Explicit context tier for the selected model. "default" / "long_context" apply the requested tier; omit this field to use normal model behavior with no explicit tier.

defer_if_model_change_queued: Option<bool>

When true, defer this switch (enqueue it) if another model change is already queued, even when no turn is active — so it drains last (FIFO) and wins over the already-queued change. Intended for genuine user-initiated model selections; internal restore/reapply switches omit it and apply immediately when no turn is active. When no other model change is queued this has no effect (a switch still applies immediately unless a turn is active).

model_capabilities: Option<ModelCapabilitiesOverride>

Override individual model capabilities resolved by the runtime

model_change_scope: Option<String>

Settings scope used when persisting the selected model.

model_id: String

Model selection id to switch to, as returned by list. A bare id (e.g. claude-sonnet-4.6) names a Copilot (CAPI) model; a provider-qualified id (provider/id, e.g. acme/claude-sonnet) targets a registry BYOK model.

picker_persistence: Option<ModelPickerPersistenceRequest>

Optional settings context and explicit-override flags used to persist a picker selection.

reasoning_effort: Option<String>

Reasoning effort level to use for the model. CAPI values are model-defined and validated against the selected model; BYOK providers may define additional values. “none” disables reasoning. Pass null to clear any session effort override and fall back to the model’s default. When omitted, the session’s current effort is kept.

reasoning_summary: Option<ReasoningSummary>

Reasoning summary mode to request for supported model clients

repo_scope: Option<String>

Optional repository settings scope to persist after the switch commits.

require_available: Option<bool>

Require the target to be currently available and enabled before applying the switch.

run_compaction_preflight: Option<bool>

When true, evaluate context-window compaction policy before applying the switch.

source: Option<ModelChangeSource>

Origin to record on the effective session.model_change event for trusted in-process calls. Transport SDK calls are always recorded as sdk, regardless of this value.

verbosity: Option<Verbosity>

Output verbosity level to request for supported models

Trait Implementations

impl Clone for ModelSwitchToRequest

fn clone(&self) -> Self

Returns a duplicate of the value.

fn clone_from(&mut self, source: &Self)

Performs copy-assignment from source.

impl Debug for ModelSwitchToRequest

fn fmt(&self, f: &mut Formatter<'_>) -> Result

Formats the value using the given formatter.

impl Default for ModelSwitchToRequest

fn default() -> Self

Returns the “default value” for a type.

impl<'de> Deserialize<'de> for ModelSwitchToRequest

fn deserialize<__D>(__deserializer: __D) -> Result<Self, __D::Error>

where __D: Deserializer<'de>,

Deserialize this value from the given Serde deserializer.

impl Serialize for ModelSwitchToRequest

fn serialize<__S>(&self, __serializer: __S) -> Result<__S::Ok, __S::Error>

where __S: Serializer,

Serialize this value into the given Serde serializer.