Struct ChatCompletionRequest 

pub struct ChatCompletionRequest {

    pub messages: Vec<ChatMessage>,
    pub model: String,
    pub frequency_penalty: Option<f32>,
    pub function_call: Option<FunctionCall>,
    pub functions: Option<Vec<Function>>,
    pub logit_bias: Option<HashMap<String, f32>>,
    pub logprobs: bool,
    pub max_tokens: Option<u32>,
    pub max_completion_tokens: Option<u32>,
    pub metadata: Option<HashMap<String, String>>,
    pub modalities: Option<Vec<String>>,
    pub n: Option<u32>,
    pub parallel_tool_calls: Option<bool>,
    pub presence_penalty: Option<f32>,
    pub prompt_cache_key: Option<String>,
    pub reasoning_effort: Option<String>,
    pub response_format: Option<ResponseFormat>,
    pub safety_identifier: Option<String>,
    pub seed: Option<i64>,
    pub service_tier: Option<String>,
    pub stop: Option<StringOrArray>,
    pub stream: bool,
    pub stream_options: Option<StreamOptions>,
    pub temperature: Option<f32>,
    pub tool_choice: Option<ToolChoice>,
    pub tools: Option<Vec<Tool>>,
    pub top_logprobs: Option<u32>,
    pub top_p: Option<f32>,
    pub verbosity: Option<i32>,
    pub top_k: Option<i32>,
    pub min_p: Option<f32>,
    pub min_tokens: Option<u32>,
    pub repetition_penalty: Option<f32>,
    pub regex: Option<String>,
    pub ebnf: Option<String>,
    pub stop_token_ids: Option<Vec<u32>>,
    pub no_stop_trim: bool,
    pub ignore_eos: bool,
    pub continue_final_message: bool,
    pub skip_special_tokens: bool,
    pub lora_path: Option<String>,
    pub session_params: Option<HashMap<String, Value>>,
    pub separate_reasoning: bool,
    pub stream_reasoning: bool,
    pub chat_template_kwargs: Option<HashMap<String, Value>>,
    pub return_hidden_states: bool,
    pub sampling_seed: Option<u64>,
}

Fields

messages: Vec<ChatMessage>

A list of messages comprising the conversation so far

model: String

ID of the model to use

frequency_penalty: Option<f32>

Number between -2.0 and 2.0. Positive values penalize new tokens based on their existing frequency in the text so far

function_call: Option<FunctionCall>

👎Deprecated: Use tool_choice instead

Deprecated: Replaced by tool_choice

functions: Option<Vec<Function>>

👎Deprecated: Use tools instead

Deprecated: Replaced by tools

logit_bias: Option<HashMap<String, f32>>

Modify the likelihood of specified tokens appearing in the completion

logprobs: bool

Whether to return log probabilities of the output tokens

max_tokens: Option<u32>

👎Deprecated: Use max_completion_tokens instead

Deprecated: Replaced by max_completion_tokens

max_completion_tokens: Option<u32>

An upper bound for the number of tokens that can be generated for a completion

metadata: Option<HashMap<String, String>>

Developer-defined tags and values used for filtering completions in the dashboard

modalities: Option<Vec<String>>

Output types that you would like the model to generate for this request

n: Option<u32>

How many chat completion choices to generate for each input message

parallel_tool_calls: Option<bool>

Whether to enable parallel function calling during tool use

presence_penalty: Option<f32>

Number between -2.0 and 2.0. Positive values penalize new tokens based on whether they appear in the text so far

prompt_cache_key: Option<String>

Cache key for prompts (beta feature)

reasoning_effort: Option<String>

Effort level for reasoning models (low, medium, high)

response_format: Option<ResponseFormat>

An object specifying the format that the model must output

safety_identifier: Option<String>

Safety identifier for content moderation

seed: Option<i64>

👎Deprecated: This feature is in Legacy mode

Deprecated: This feature is in Legacy mode

service_tier: Option<String>

The service tier to use for this request

stop: Option<StringOrArray>

Up to 4 sequences where the API will stop generating further tokens

stream: bool

If set, partial message deltas will be sent

stream_options: Option<StreamOptions>

Options for streaming response

temperature: Option<f32>

What sampling temperature to use, between 0 and 2

tool_choice: Option<ToolChoice>

Controls which (if any) tool is called by the model

tools: Option<Vec<Tool>>

A list of tools the model may call

top_logprobs: Option<u32>

An integer between 0 and 20 specifying the number of most likely tokens to return

top_p: Option<f32>

An alternative to sampling with temperature

verbosity: Option<i32>

Verbosity level for debugging

top_k: Option<i32>

Top-k sampling parameter (-1 to disable)

min_p: Option<f32>

Min-p nucleus sampling parameter

min_tokens: Option<u32>

Minimum number of tokens to generate

repetition_penalty: Option<f32>

Repetition penalty for reducing repetitive text

regex: Option<String>

Regex constraint for output generation

ebnf: Option<String>

EBNF grammar constraint for structured output

stop_token_ids: Option<Vec<u32>>

Specific token IDs to use as stop conditions

no_stop_trim: bool

Skip trimming stop tokens from output

ignore_eos: bool

Ignore end-of-sequence tokens during generation

continue_final_message: bool

Continue generating from final assistant message

skip_special_tokens: bool

Skip special tokens during detokenization

lora_path: Option<String>

Path to LoRA adapter(s) for model customization

session_params: Option<HashMap<String, Value>>

Session parameters for continual prompting

separate_reasoning: bool

Separate reasoning content from final answer (O1-style models)

stream_reasoning: bool

Stream reasoning tokens during generation

chat_template_kwargs: Option<HashMap<String, Value>>

Chat template kwargs

return_hidden_states: bool

Return model hidden states

sampling_seed: Option<u64>

Random seed for sampling for deterministic outputs

Trait Implementations

impl Clone for ChatCompletionRequest

fn clone(&self) -> ChatCompletionRequest

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for ChatCompletionRequest

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

Formats the value using the given formatter.

impl Default for ChatCompletionRequest

fn default() -> ChatCompletionRequest

Returns the “default value” for a type.

impl<'de> Deserialize<'de> for ChatCompletionRequest

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

where __D: Deserializer<'de>,

Deserialize this value from the given Serde deserializer.

impl GenerationRequest for ChatCompletionRequest

fn is_stream(&self) -> bool

Check if the request is for streaming

fn get_model(&self) -> Option<&str>

Get the model name if specified

fn extract_text_for_routing(&self) -> String

Extract text content for routing decisions

impl Normalizable for ChatCompletionRequest

fn normalize(&mut self)

Normalize the request by applying migrations and defaults:

1. Migrate deprecated fields to their replacements
2. Clear deprecated fields and log warnings
3. Apply OpenAI defaults for tool_choice

impl Serialize for ChatCompletionRequest

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

where __S: Serializer,

Serialize this value into the given Serde serializer.

impl Validate for ChatCompletionRequest

fn validate(&self) -> Result<(), ValidationErrors>

impl<'v_a> ValidateArgs<'v_a> for ChatCompletionRequest

type Args = ()

fn validate_with_args(&self, args: Self::Args) -> Result<(), ValidationErrors>