pub trait OAIChatLikeRequest {
Show 13 methods
// Required methods
fn model(&self) -> String;
fn messages(&self) -> Value;
fn should_add_generation_prompt(&self) -> bool;
// Provided methods
fn typed_messages(&self) -> Option<&[ChatCompletionRequestMessage]> { ... }
fn tools(&self) -> Option<Value> { ... }
fn tool_choice(&self) -> Option<Value> { ... }
fn response_format(&self) -> Option<Value> { ... }
fn reasoning_effort(&self) -> Option<Value> { ... }
fn chat_template_args(&self) -> Option<&HashMap<String, Value>> { ... }
fn prompt_input_type(&self) -> PromptInput { ... }
fn extract_tokens(&self) -> Option<TokenInput> { ... }
fn extract_text(&self) -> Option<TextInput> { ... }
fn mm_processor_kwargs(&self) -> Option<&Value> { ... }
}Expand description
Trait that defines a request that can map to an OpenAI-like request.
Implement this for your request type to render it through a
PromptFormatter. Media/multimodal IO config is intentionally not part
of this trait — it is a preprocessing concern owned by the consumer, kept
off the rendering surface so this crate stays runtime-free.
Required Methods§
fn model(&self) -> String
fn messages(&self) -> Value
fn should_add_generation_prompt(&self) -> bool
Provided Methods§
fn typed_messages(&self) -> Option<&[ChatCompletionRequestMessage]>
fn tools(&self) -> Option<Value>
fn tool_choice(&self) -> Option<Value>
fn response_format(&self) -> Option<Value>
Sourcefn reasoning_effort(&self) -> Option<Value>
fn reasoning_effort(&self) -> Option<Value>
OpenAI-compatible reasoning-effort control, when the request type exposes it as a top-level field.
Sourcefn chat_template_args(&self) -> Option<&HashMap<String, Value>>
fn chat_template_args(&self) -> Option<&HashMap<String, Value>>
Optional additional args to merge into the chat template context
Sourcefn prompt_input_type(&self) -> PromptInput
fn prompt_input_type(&self) -> PromptInput
Returns the type of input for the prompt. Default is Text.
Sourcefn extract_tokens(&self) -> Option<TokenInput>
fn extract_tokens(&self) -> Option<TokenInput>
Extract tokens if the input is pre-tokenized
fn extract_text(&self) -> Option<TextInput>
fn mm_processor_kwargs(&self) -> Option<&Value>
Dyn Compatibility§
This trait is dyn compatible.
In older versions of Rust, dyn compatibility was called "object safety".
Implementations on Foreign Types§
Source§impl OAIChatLikeRequest for CreateChatCompletionRequest
Default OAIChatLikeRequest impl for the bare dynamo-protocols chat
request. Lets any consumer (e.g. a standalone OpenAI frontend over an
engine) render HF chat templates directly from the wire type, without
defining their own wrapper. Consumers with extra fields (Dynamo’s
NvCreateChatCompletionRequest) provide their own impl.
impl OAIChatLikeRequest for CreateChatCompletionRequest
Default OAIChatLikeRequest impl for the bare dynamo-protocols chat
request. Lets any consumer (e.g. a standalone OpenAI frontend over an
engine) render HF chat templates directly from the wire type, without
defining their own wrapper. Consumers with extra fields (Dynamo’s
NvCreateChatCompletionRequest) provide their own impl.