Struct OpenAIProvider 

pub struct OpenAIProvider { /* private fields */ }

OpenAI LLM provider using the Chat Completions API.

Also supports OpenAI-compatible APIs (Ollama, vLLM, Azure OpenAI, etc.) via the with_base_url constructor.

Implementations

impl OpenAIProvider

pub fn new(api_key: String, model: String) -> Self

Create a new OpenAI provider with the specified API key and model.

pub fn with_base_url(api_key: String, model: String, base_url: String) -> Self

Create a new provider with a custom base URL for OpenAI-compatible APIs.

pub fn kimi(api_key: String, model: String) -> Self

Create a provider using Moonshot KIMI via OpenAI-compatible Chat Completions.

pub fn kimi_k2_5(api_key: String) -> Self

Create a provider using KIMI K2.5 (default KIMI model).

pub fn kimi_k2_thinking(api_key: String) -> Self

Create a provider using KIMI K2 Thinking.

pub fn zai(api_key: String, model: String) -> Self

Create a provider using z.ai via OpenAI-compatible Chat Completions.

pub fn zai_glm5(api_key: String) -> Self

Create a provider using z.ai GLM-5 (default z.ai agentic reasoning model).

pub fn minimax(api_key: String, model: String) -> Self

Create a provider using MiniMax via OpenAI-compatible Chat Completions.

pub fn minimax_m2_5(api_key: String) -> Self

Create a provider using MiniMax M2.5 (default MiniMax model).

pub fn gpt52_instant(api_key: String) -> Self

Create a provider using GPT-5.2 Instant (speed-optimized for routine queries).

pub fn gpt54(api_key: String) -> Self

Create a provider using GPT-5.4 (frontier reasoning with 1.05M context).

pub fn gpt53_codex(api_key: String) -> Self

Create a provider using GPT-5.3 Codex (latest codex model).

pub fn gpt52_thinking(api_key: String) -> Self

Create a provider using GPT-5.2 Thinking (complex reasoning, coding, analysis).

pub fn gpt52_pro(api_key: String) -> Self

Create a provider using GPT-5.2 Pro (maximum accuracy for difficult problems).

pub fn codex(api_key: String) -> Self

Create a provider using the latest Codex model.

pub fn gpt5(api_key: String) -> Self

Create a provider using GPT-5 (400k context, coding and reasoning).

pub fn gpt5_mini(api_key: String) -> Self

Create a provider using GPT-5-mini (faster, cost-efficient GPT-5).

pub fn gpt5_nano(api_key: String) -> Self

Create a provider using GPT-5-nano (fastest, cheapest GPT-5 variant).

pub fn o3(api_key: String) -> Self

Create a provider using o3 (most intelligent reasoning model).

pub fn o3_mini(api_key: String) -> Self

Create a provider using o3-mini (smaller o3 variant).

pub fn o4_mini(api_key: String) -> Self

Create a provider using o4-mini (fast, cost-efficient reasoning).

pub fn o1(api_key: String) -> Self

Create a provider using o1 (reasoning model).

pub fn o1_mini(api_key: String) -> Self

Create a provider using o1-mini (fast reasoning model).

pub fn gpt41(api_key: String) -> Self

Create a provider using GPT-4.1 (improved instruction following, 1M context).

pub fn gpt41_mini(api_key: String) -> Self

Create a provider using GPT-4.1-mini (smaller, faster GPT-4.1).

pub fn gpt4o(api_key: String) -> Self

Create a provider using GPT-4o.

pub fn gpt4o_mini(api_key: String) -> Self

Create a provider using GPT-4o-mini (fast and cost-effective).

pub const fn with_thinking(self, thinking: ThinkingConfig) -> Self

Set the provider-owned thinking configuration for this model.

Trait Implementations

impl Clone for OpenAIProvider

fn clone(&self) -> OpenAIProvider

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl LlmProvider for OpenAIProvider

fn chat<'life0, 'async_trait>( &'life0 self, request: ChatRequest, ) -> Pin<Box<dyn Future<Output = Result<ChatOutcome>> + Send + 'async_trait>>

where Self: 'async_trait, 'life0: 'async_trait,

Non-streaming chat completion.

fn chat_stream(&self, request: ChatRequest) -> StreamBox<'_>

Streaming chat completion.

fn model(&self) -> &str

fn provider(&self) -> &'static str

fn configured_thinking(&self) -> Option<&ThinkingConfig>

Provider-owned thinking configuration, if any.

fn capabilities(&self) -> Option<&'static ModelCapabilities>

Canonical capability metadata for this provider/model, if known.

fn validate_thinking_config( &self, thinking: Option<&ThinkingConfig>, ) -> Result<()>

Validate a thinking configuration against the provider/model capabilities.

fn resolve_thinking_config( &self, request_thinking: Option<&ThinkingConfig>, ) -> Result<Option<ThinkingConfig>>

Resolve the effective thinking configuration for a request.

fn default_max_tokens(&self) -> u32

Default maximum output tokens for this provider/model when the caller does not explicitly override AgentConfig.max_tokens.