Enum Thinking 

#[non_exhaustive]
pub enum Thinking {
    Unset,
    Off,
    On(Effort),
}

Whether the model should reason before answering.

Three states rather than a bool, and the third is the one that matters. On some models reasoning is on by default and on others it is off, so collapsing “no opinion” into “off” silently changes behaviour the moment a model id changes.

Whether a model can reason is a different question, answered by crate::ModelCapabilities::thinking. It is deliberately not a variant here: a request says what you want, a capability says what is possible, and one enum carrying both would be two answers to two questions in one place.

Variants (Non-exhaustive)

Non-exhaustive enums could have additional variants added in future. Therefore, when matching against variants of non-exhaustive enums, an extra wildcard arm must be added to account for any future variants.

Unset

No opinion. Whatever the model does by default.

Off

Do not reason.

On(Effort)

Reason at this level.

Trait Implementations

impl Clone for Thinking

fn clone(&self) -> Thinking

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Copy for Thinking

impl Debug for Thinking

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

Formats the value using the given formatter.

impl Default for Thinking

fn default() -> Thinking

Returns the “default value” for a type.

impl<'de> Deserialize<'de> for Thinking

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

where __D: Deserializer<'de>,

Deserialize this value from the given Serde deserializer.

impl Eq for Thinking

impl PartialEq for Thinking

fn eq(&self, other: &Thinking) -> bool

Equality operator ==.

fn ne(&self, other: &Rhs) -> bool

Inequality operator !=.

impl Serialize for Thinking

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

where __S: Serializer,

Serialize this value into the given Serde serializer.

impl StructuralPartialEq for Thinking