Enum ReasoningStrategy 

pub enum ReasoningStrategy {
    None,
    Effort,
    OpenRouterShape,
}

How to put ReasoningLevel onto the wire for a given provider.

Variants

None

Provider exposes no reasoning controls (Together, DeepInfra pass-through). Adapter sends nothing extra.

Effort

Standard reasoning_effort: "low"|"medium"|"high" field (OpenAI Chat Completions, Groq for gpt-oss, Cerebras for gpt-oss-120b, Fireworks for Qwen 3, etc.).

OpenRouterShape

OpenRouter’s normalized reasoning: {effort: "..."} nested object. Supports low, medium, high, max. None becomes {exclude: true} (suppresses reasoning).

Implementations

impl ReasoningStrategy

pub fn render(&self, level: ReasoningLevel) -> Option<Value>

Render a ReasoningLevel to the JSON fragment that should be merged into the /chat/completions request body. Returns None if there’s nothing to add (strategy is None, or the level is None for a provider that signals via field omission).

Trait Implementations

impl Clone for ReasoningStrategy

fn clone(&self) -> ReasoningStrategy

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Copy for ReasoningStrategy

impl Debug for ReasoningStrategy

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

Formats the value using the given formatter.

impl Eq for ReasoningStrategy

impl PartialEq for ReasoningStrategy

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

Tests for self and other values to be equal, and is used by ==.

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

Tests for !=. The default implementation is almost always sufficient, and should not be overridden without very good reason.

impl StructuralPartialEq for ReasoningStrategy