Struct Model 

pub struct Model {

    pub name: &'static str,
    pub family: &'static str,
    pub version: &'static str,
    pub anthropic_id: &'static str,
    pub bedrock_id: Option<&'static str>,
    pub bedrock_global_id: Option<&'static str>,
    pub vertex_id: Option<&'static str>,
    pub max_context_tokens: u32,
    pub max_context_tokens_extended: Option<u32>,
    pub max_output_tokens: u32,
    pub supports_vision: bool,
    pub supports_tools: bool,
    pub supports_caching: bool,
    pub supports_extended_thinking: bool,
    pub supports_effort: bool,
    pub cost_per_mtok_input: f64,
    pub cost_per_mtok_output: f64,
    pub description: &'static str,
}

Model capabilities and constraints

Fields

name: &'static str

Human-readable model name

family: &'static str

Model family (e.g., “sonnet”, “opus”, “haiku”)

version: &'static str

Release date/version identifier

anthropic_id: &'static str

Anthropic API model identifier

bedrock_id: Option<&'static str>

AWS Bedrock regional endpoint model identifier (if available)

bedrock_global_id: Option<&'static str>

AWS Bedrock global endpoint model identifier (if available)

Claude 4.5+ models support global endpoints for dynamic routing. Use this for maximum availability across regions.

vertex_id: Option<&'static str>

Google Vertex AI model identifier (if available)

max_context_tokens: u32

Maximum context window in tokens (standard)

max_context_tokens_extended: Option<u32>

Extended context window in tokens (if available)

Some models support extended context with beta headers. For example, Claude Sonnet 4.5 and Sonnet 4 support 1M tokens with the context-1m-2025-08-07 beta header.

max_output_tokens: u32

Maximum output tokens per request

supports_vision: bool

Supports vision (image inputs)

supports_tools: bool

Supports tool use

supports_caching: bool

Supports prompt caching

supports_extended_thinking: bool

Supports extended thinking

supports_effort: bool

Supports effort parameter (beta)

Requires beta header: anthropic-beta: effort-2025-11-24 Currently only Claude Opus 4.5

cost_per_mtok_input: f64

Cost per million input tokens (USD)

cost_per_mtok_output: f64

Cost per million output tokens (USD)

description: &'static str

Brief description of best use cases

Implementations

impl Model

pub fn anthropic_id(&self) -> &'static str

Get the model ID for the Anthropic API

pub fn bedrock_id(&self) -> Option<&'static str>

Get the model ID for AWS Bedrock regional endpoint (if available)

pub fn bedrock_id_for_region(&self, region: BedrockRegion) -> Option<String>

Get the model ID for AWS Bedrock with a specific region prefix

Example

use claude_sdk::models::{CLAUDE_SONNET_4_5, BedrockRegion};

// Standard regional endpoint
let regional = CLAUDE_SONNET_4_5.bedrock_id_for_region(BedrockRegion::Standard);
// → Some("anthropic.claude-sonnet-4-5-20250929-v1:0")

// Global endpoint
let global = CLAUDE_SONNET_4_5.bedrock_id_for_region(BedrockRegion::Global);
// → Some("global.anthropic.claude-sonnet-4-5-20250929-v1:0")

// US regional endpoint
let us = CLAUDE_SONNET_4_5.bedrock_id_for_region(BedrockRegion::US);
// → Some("us.anthropic.claude-sonnet-4-5-20250929-v1:0")

pub fn bedrock_global_id(&self) -> Option<&'static str>

Get the model ID for AWS Bedrock global endpoint (if available)

Global endpoints provide dynamic routing for maximum availability. Available for Claude 4.5+ models.

This is a convenience method equivalent to bedrock_id_for_region(BedrockRegion::Global).

pub fn vertex_id(&self) -> Option<&'static str>

Get the model ID for Google Vertex AI (if available)

pub fn supports_extended_context(&self) -> bool

Check if this model supports extended context (e.g., 1M tokens)

pub fn max_extended_context(&self) -> Option<u32>

Get the extended context window size (if supported)

Returns Some(tokens) if the model supports extended context with beta headers. For example, Claude Sonnet 4.5 returns Some(1_000_000).

Beta Header Required

To use extended context, include the beta header in your API request:

• Header: anthropic-beta: context-1m-2025-08-07

Note: Extended context may incur additional costs beyond 200K tokens.

pub fn validate_request( &self, max_tokens: u32, use_extended_context: bool, ) -> Result<(), String>

Validate that a request is compatible with this model’s constraints

Parameters

• max_tokens: The requested maximum output tokens
• use_extended_context: Whether extended context will be used

pub fn estimate_cost(&self, input_tokens: u32, output_tokens: u32) -> f64

Estimate cost for a request

Trait Implementations

impl Clone for Model

fn clone(&self) -> Model

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for Model

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

Formats the value using the given formatter.

impl PartialEq for Model

fn eq(&self, other: &Model) -> 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 Model