Module request 

Inference request management for AI model providers.

Provides a minimal blocking interface for sending prompts via configured Providers and returning a unified InferenceResponse. Supports OpenAI-compatible and Ollama chat APIs using the Prompt abstraction.

Structs

HttpResponse

Internal, provider-agnostic HTTP response used by the inference client abstraction.

InferenceResponse

Unified response for a single provider request, including text and token usage.

OllamaChatResponse

Response structure from Ollama chat API

OllamaMessage

OllamaMessageResponse

OllamaOptions

OpenAiChoice

OpenAiMessage

OpenAiRequest

OpenAiResponse

OpenAiUsage

ReqwestInferenceClient

Default blocking HTTP client implementation backed by reqwest.

Traits

InferenceHttpClient

Minimal HTTP client abstraction for inference requests.

Functions

send_request

Public entry point: send a request using the default blocking HTTP client.

send_request_with_client

Internal helper: core logic parameterized over an InferenceHttpClient.