Struct ModelCapabilities 

pub struct ModelCapabilities {
    pub context_length: usize,
    pub max_output_tokens: usize,
    pub supports_tools: bool,
    pub supports_vision: bool,
    pub supports_streaming: bool,
    pub provider: String,
    pub model_id: String,
    pub source: CapabilitySource,
}

Model capabilities describing limits and supported features.

Fields

context_length: usize

Maximum context window in tokens, always stored as input + output combined regardless of whether the provider exposes a single combined limit (e.g. OpenAI) or separate per-modality limits (e.g. Anthropic max_input_tokens, Gemini inputTokenLimit).

Do not use this field directly for prompt-budget decisions. Always call Self::max_input_tokens() instead, which subtracts max_output_tokens to yield the tokens available for the prompt.

Invariant

context_length = max_input_tokens() + max_output_tokens

Discovery code for providers with separate input/output limits must store input_limit + output_limit here so that max_input_tokens() correctly recovers input_limit without double-subtracting.

max_output_tokens: usize

Maximum output/completion tokens the model can generate

supports_tools: bool

Whether the model supports native tool/function calling

supports_vision: bool

Whether the model supports vision/image inputs

supports_streaming: bool

Whether the model supports streaming responses

provider: String

Provider name for reference

model_id: String

Model ID for reference

source: CapabilitySource

How this capability was discovered

Implementations

impl ModelCapabilities

pub fn new( provider: &str, model_id: &str, context_length: usize, max_output_tokens: usize, source: CapabilitySource, ) -> Self

Create capabilities with known values

pub fn with_vision(self, supports: bool) -> Self

Set vision support

pub fn with_tools(self, supports: bool) -> Self

Set tool support

pub fn max_input_tokens(&self) -> usize

Calculate the effective max input tokens (context - reserved output)

pub fn is_reliable(&self) -> bool

Check if this capability is from a reliable source

Trait Implementations

impl Clone for ModelCapabilities

fn clone(&self) -> ModelCapabilities

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for ModelCapabilities

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

Formats the value using the given formatter.

impl Default for ModelCapabilities

fn default() -> Self

Returns the “default value” for a type.

impl<'de> Deserialize<'de> for ModelCapabilities

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

where __D: Deserializer<'de>,

Deserialize this value from the given Serde deserializer.

impl PartialEq for ModelCapabilities

fn eq(&self, other: &ModelCapabilities) -> 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 Serialize for ModelCapabilities

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

where __S: Serializer,

Serialize this value into the given Serde serializer.

impl Eq for ModelCapabilities

impl StructuralPartialEq for ModelCapabilities