API Endpoints

Endpoint List

smai.ai provides the following API endpoints:

Endpoint	Purpose	URL
Standard Endpoint	General API base address	`https://api.smai.ai/v1`
Chat Completions	Conversation completion	`https://api.smai.ai/v1/chat/completions`
Responses	Inference model specific	`https://api.smai.ai/v1/responses`
Models	Get list of models	`https://api.smai.ai/v1/models`

Standard Endpoint

https://api.smai.ai/v1

This is the base address for the API, which should be configured when using the SDK.

from openai import OpenAI

client = OpenAI(
    api_key="sk-your-api-key",
    base_url="https://api.smai.ai/v1"  # Standard Endpoint
)

Chat Completions Endpoint

https://api.smai.ai/v1/chat/completions

Used for conversation completion, supporting most models. See Chat Completions API for details.

Responses Endpoint

https://api.smai.ai/v1/responses

Inference models must use this endpoint

The following models must use the Responses endpoint:

gpt-5.2-pro
o3-pro
o3-mini
Other inference models

See Responses API for details.

Models Endpoint

https://api.smai.ai/v1/models

Get the list of currently available models:

curl https://api.smai.ai/v1/models \
  -H "Authorization: Bearer sk-your-api-key"

Base Endpoint

https://api.smai.ai

The base address without a version number, which some applications may need to use.

Endpoint Selection Guide

graph TD
    A[Select Endpoint] --> B{Is it an inference model?}
    B -->|Yes| C[Use Responses Endpoint]
    B -->|No| D[Use Chat Completions Endpoint]
    C --> E[/v1/responses]
    D --> F[/v1/chat/completions]

Chat Completions API

Detailed explanation of the conversation completion interface

Responses API

Detailed explanation of the inference model interface